Slashdot Mirror


Sun CEO Says ZFS Will Be 'the File System' for OSX

Fjan11 writes "Sun's Jonathan Schwartz has announced that Apple will be making ZFS 'the file system' in Mac OS 10.5 Leopard. It's possible that Leopard's Time Machine feature will require ZFS to run, because ZFS has back-up and snapshots build right in to the filesystem as well as a host of other features. 'Rumors of Apple's interest in ZFS began in April 2006, when an OpenSolaris mailing list revealed that Apple had contacted Sun regarding porting ZFS to OS 10. The file system later began making appearances in Leopard builds. ZFS has a long list of improvements over Apple's current file system, Journaled HFS+.'"

384 comments

  1. oblig... by doyoulikeworms · · Score: 0, Offtopic

    One file system to rule them all.

    1. Re:oblig... by eln · · Score: 3, Interesting

      We had a guy come in a few months ago to give a class on upgrading to Solaris 10, highlighting the differences between Solaris 9 and 10. When he got to the ZFS portion, he really did talk about it like that. He basically described ZFS as the filesystem to end all filesystems, the killer app that would revolutionize computing, end file corruption, and bring about world peace.

      I'm not sure if that's the way they talk about it internally at Sun, but that's how their instructors portray it out in the field.

    2. Re:oblig... by kildurin · · Score: 5, Informative

      Its worth noting that most Sun instructors do not work for Sun. As someone who has implemented and is using ZFS, it really is as good as they say. I use it at home for storing video files and have not suffered any data loss.

    3. Re:oblig... by djh101010 · · Score: 3, Interesting

      We had a guy come in a few months ago to give a class on upgrading to Solaris 10, highlighting the differences between Solaris 9 and 10. When he got to the ZFS portion, he really did talk about it like that. He basically described ZFS as the filesystem to end all filesystems, the killer app that would revolutionize computing, end file corruption, and bring about world peace.
      That's quite a change from about a year ago, when I took the "new features in Solaris10" class; at that time the instructor I had was in no uncertain terms saying it's "not ready for production, wait until later". Apparently we have reached "later"? Or it could be that people have opinions and express them, and aren't all speaking for Sun; I suppose that's possible...
    4. Re:oblig... by kildurin · · Score: 3, Informative

      ZFS came out in Solaris 10 Update 2. (Sun is days away from releasing update 4.) It is currently bootable in OpenSolaris.

    5. Re:oblig... by Anarke_Incarnate · · Score: 3, Informative

      It still has memory hogging issues as well as performance issues in certain areas. More kernel tuning will be needed to tame the beast that is ZFS. It is good for many things but it does not replace EVERYTHING just yet.

    6. Re:oblig... by BosstonesOwn · · Score: 5, Interesting

      as a worker at sun and having used ZFS and playing with it constantly , it is a good File system , I appreciate the little things it has and it has brought data stability to a whole new level. I think personally that this will be a defining moment for ZFS , it will be linux ready soon ( at the same level of stability that the mac will enjoy ) and it will take off and become more of a standard for unix and linux boxes.

      To bad no windows port is available. It would be nice to see my unix drives from windows.

      --
      This package Does Not Contain a Winner
    7. Re:oblig... by goombah99 · · Score: 2, Interesting

      One file system to rule them all. I thought that was Plan-9. If they really want to look ahead why not look to plan-9
      --
      Some drink at the fountain of knowledge. Others just gargle.
    8. Re:oblig... by evil_Tak · · Score: 3, Funny

      Apparently you've never used ReiserFS.

    9. Re:oblig... by mapinguari · · Score: 4, Funny

      You misunderstood him. He said that it would bring about "War and Peace", not world peace.
      Every ZFS volume has a copy of the Tolstoy classic embedded for internal benchmark purposes.

    10. Re:oblig... by Anonymous Coward · · Score: 5, Funny

      He wrote 'data loss', not 'wife loss'.

    11. Re:oblig... by MrHanky · · Score: 0, Offtopic

      My /home partition on my main computer is ReiserFS, and has been since at least 2003. Never had problems with it. I've stored video files on that one as well. In fact, the only FS I've lost files with recently is HFS+ on my old Powerbook. I'm just saying that not having problems is what's expected. FAT32 is the poorest FS still in use, and normally you won't lose data even with that.

    12. Re:oblig... by morgan_greywolf · · Score: 2, Interesting

      it will be linux ready soon


      Aren't there still licensing issues to iron out?

      To bad no windows port is available. It would be nice to see my unix drives from windows.


      Ext2/3 and ReiserFS have all been ported to Windows, so I don't see there would be any problem porting the Linux ZFS implementation as an IFS driver for Windows Vista/XP/2000

    13. Re:oblig... by evil_Tak · · Score: 2, Interesting

      I must have some nonstandard use case, because I've lost data with both ReiserFS (multiple times, as I optimistically tried newer versions) and XFS.

      Ext3, on the other hand, has been rock solid for me.

    14. Re:oblig... by rainman_bc · · Score: 3, Interesting

      it will be linux ready soon ( at the same level of stability that the mac will enjoy ) and it will take off and become more of a standard for unix and linux boxes.

      Depends - whatcha building? An app server, a web server, a database server, or a file server? Different strokes for different folks, and I'm not clear yet if I'd like the overhead of ZFS on a database server. The jury's still out on ZFS+Oracle...

      I'm not 100% on which file system I'd like. Certainly the integrity of ZFS is quite pleasing for a DBA, questions is if the overhead is worth it...

      --
      09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
    15. Re:oblig... by someone300 · · Score: 0, Offtopic

      My ReiserFS partitions are fine, as are my ext2/3 partitions and my ZFS partitions. My HFS+ partitions, on the other hand, are the least reliable for some odd reason, regardless of what hard drive, computer or operating system (OS X, Linux) I've tried. Even safely shutting down and dismounting. On OS X at least, it seems to corrupt under normal use for me, on my iPod and my internal drives. Very strange.

      Hell, the only time I really lose data with FAT32 in my memory was when I pull out an in-use hard drive or crash the system.

    16. Re:oblig... by Anonymous Coward · · Score: 3, Funny

      I married Brent Spiner, you insensitive clod.

    17. Re:oblig... by piojo · · Score: 0, Offtopic

      You are not alone--with slackware 10.0 and the older version of reiserfs that came with it, I repeatedly suffered filesystem corruption. I couldn't comprehend how they shipped such a buggy FS, but true linux geeks might not discover any problem--it only happened when I unzipped windows archives.

      --
      A cat can't teach a dog to bark.
    18. Re:oblig... by BosstonesOwn · · Score: 3, Informative

      On linux FUSE is all we can use because of the license restrictions. I use it on a Solaris box with a vmed red hat install it's not quite ready for prime time. And the licensing sucks because it won't get put into the kernel.

      Which I find really deplurable , I would love to have it in the kernel.

      And yeah id like an ifs driver but none exist ;) yet and I don't see any intent internally to even attempt it. as it stand I think I am maybe 1 of 8 people using ZFS on linux here. And that is only in the VM.

      --
      This package Does Not Contain a Winner
    19. Re:oblig... by dougmc · · Score: 2, Insightful

      I use it at home for storing video files and have not suffered any data loss. The plural of anecdote is not data.

      ZFS does indeed look like the greatest thing since sliced bread, but if all I cared about `wasn't losing data', then I'd just stick with ext2 or fat. (But I care about more than that, which is why I use xfs, ext3, hfs+ and ntfs for various things.)

    20. Re:oblig... by Anarke_Incarnate · · Score: 1

      With enough spindles you could use it on a DB server if you set the max ARC size to a smaller number to keep the cache from growing to insane sizes. You would also want to create the volumes with the write block size that the database uses for random reads (usually 8K) and also set the read size to the same value. Then you should be able to get performance similar to other file systems like UFS. You would lose some of the features that you would probably not care about. Realize ZFS has not been "out that long" relative to many other stable file systems, and as such a huge undertaking, it is going to have major bugs to squash. I think it was released well before it was primetime, but it is getting better.

    21. Re:oblig... by Anonymous Coward · · Score: 0

      ...as a worker at sun and having used ZFS and playing with it constantly , it is a good File system... Yeah, but now that Apple is using it ZFS has gone from being merely good to being 'cool'. Imagine that... a 'cool' product... from Sun!!!
    22. Re:oblig... by moosesocks · · Score: 3, Interesting

      From what I understand, the API that Windows filesystem drivers interface with is an undocumented nightmare that's entirely different (but not necessarily 'worse') from the way the rest of the world does it.

      So porting a filesystem as complicated as ZFS could take some time.

      --
      -- If you try to fail and succeed, which have you done? - Uli's moose
    23. Re:oblig... by Angus+McNitt · · Score: 2, Funny

      Shh, that is an inside secret. They were originally going to use The Brothers Karamazov, but the developers couldn't agree on a pronunciation.

      --
      "To Do Is To Be" - Socrates, "To Be Is To Do" - Sartre, "Do Be Do Be Do" - Sinatra
    24. Re:oblig... by EugeneK · · Score: 0, Offtopic

      Just another person piling on to say that I've use Reiser FS for at least the past 2 years, as far as I can remember, and have had no problems with it.

    25. Re:oblig... by Cyberax · · Score: 4, Informative

      Windows filesystem kernel API (it's called IFS - Installable File Systems) is fairly well documented, and you can get free GPL2 headers for it (http://www.acc.umu.se/~bosse/ntifs.h) or buy IFS kit from Microsoft for about $109 (http://www.microsoft.com/whdc/DevTools/IFSKit/def ault.mspx). Unfortunately, IFS is a very complex API and there's only ONE good book about it.

      You definitely can port FS to Windows using only documented API, but it's a long and tedious process. I'm currently porting FUSE to Windows, so I know it :)

    26. Re:oblig... by Anonymous Coward · · Score: 0

      I thought it was "whirled peas"?

    27. Re:oblig... by Anonymous Coward · · Score: 2, Informative

      there are no licensing issues with linux/zfs. The "issue" is that Alan Cox and Linus Torvaldes have a hot iron up their asses about compartmentalizing everything. ZFS is vertical -- it's in the fs layer, the driver layer, and the vfs layer.

    28. Re:oblig... by adisakp · · Score: 5, Funny

      Nope. ZFS WAS going to be the file system until Sun's CEO leaked it before Jobs could announce it. No one knows what it will actually be NOW. Jobs would rather use FAT32 than let someone leak info about one of his announcements.

    29. Re:oblig... by Anonymous Coward · · Score: 0

      MODS, why is this a troll? if anything, the parent is way over-rated.

    30. Re:oblig... by MrHanky · · Score: 1

      It's not a troll, of course, and not even off topic. But I accidentally bad-mouthed the sacred HFS+ in a later comment, which is a mortal sin to Mac fanboys.

    31. Re:oblig... by Tribbin · · Score: 3, Funny

      That was user error.

      --
      If you mod this up, your slashdot background will turn into a beautiful sunset!
    32. Re:oblig... by Anonymous Coward · · Score: 1, Funny

      >Which I find really deplurable
      Does that mean you find it singular?

    33. Re:oblig... by salimma · · Score: 2, Informative

      Yes there is. ZFS is licensed under CDDL, which is not GPL2 compatible. Linus has so far refused to move to GPL3 when it comes out, so before that happens, there is a licensing issue.

      --
      Michel
      Fedora Project Contribut
    34. Re:oblig... by mink · · Score: 1

      I hope it continues to improve on Linux. I am running it on slackware 11 via zfs-fuse.

      I have a A1000 Storage array populated with 34GB drives. The only SUN box I have is an old SPARCserver chassis with 1 main board (2 procs) and a SPARCstation 20 (2 procs). It cant handle Solaris above version 8. So I took it to x86 land.

        I managed to get Linux to see these drives (LASI HBA) after low leveling them to a 512 sector size (they were salvaged from arrays with 520 and 528 sector sizes). I decided to try setting up a raidz2 across 12 of them and set the last 2 as spares this is giving me 390GB of storage. I made them accessible with samba and am dumping a few gigs of video onto them as a torture test.

      So far I think it is working well, but I think I need to tune samba or give the machine a bigger processor or more memory because it seems to be taking a lot longer to dump 180GB onto it then I expected. It currently has 512 MB RAM and an old 1.33 GHZ Thunderbird core Athlon.

      I find it interesting that I initially made the array with 11 disks (I missed one of the drive names during creation) and only got 300GB, while across the 12 I get 390. Much better then RAID 5 so far IMO.

      --
      Well I've wrestled with reality for thirty five years doctor, and I'm happy to say I finally won out over it.
    35. Re:oblig... by Anonymous Coward · · Score: 0
      Brought stability to a whole new level? What do you mean? How did you measure that?


      I'm a kernel hacker and I did research on filesystem reliability. I power cycle tested a half dozen filesystems 10s of thousands of times. No filesystem is perfect, at least not any that I had tested and the best all seemed to fall within the margin of noise regarding their reliability. How is ZFS any more reliable than FFS with soft updates or EXT3? Any evidence or just it feels that way?


      How does it react if you lose a volume? Being that it has no fsck? I think this is a very real concern, fscking a large zfs could take weeks.

  2. The were going to use Reiser by Anonymous Coward · · Score: 4, Funny

    But they killed that project.

    1. Re:The were going to use Reiser by dch24 · · Score: 2, Interesting

      As a linux user, I have found good use for ReiserFS. However, I've been asked time and again "why doesn't my iPod work with Windows"? If they move to an open source file system, iTunes for Windows could easily include a ZFS driver. (Correct me if I'm wrong, but some sort of ZFS driver is in the Linux kernel, and Sun is open sourcing Solaris.)

      I like having an mp3 player that doubles as a backup device for my important files. But some of my files are > 4Gb, so FAT32 doesn't work.

    2. Re:The were going to use Reiser by aled · · Score: 1, Funny

      But they killed that project.


      This is Slashdot. We do not say "they killed the project". We say "but BSD/OpenSolaris/ZFS/etc is dead".
      --

      "I think this line is mostly filler"
    3. Re:The were going to use Reiser by soleblaze · · Score: 3, Interesting

      ZFS cannot be added to the linux kernel due to licensing issues. However, there is work being done on a FUSE module for ZFS support. Though I'm not sure if it'll be worth using for anything more than accessing existing ZFS partitions.

    4. Re:The were going to use Reiser by brunascle · · Score: 4, Informative

      Correct me if I'm wrong, but some sort of ZFS driver is in the Linux kernel
      i dont think there is (could be wrong). something about a licensing problem. but apparently some people have gotten it to work in linux using FUSE. (more info)
    5. Re:The were going to use Reiser by Anonymous Coward · · Score: 4, Funny

      You know, having a project like that canceled is very stressful for developers. It must have been murder on their families, too.

      Oh God. I can't sign my name to that.

    6. Re:The were going to use Reiser by Hes+Nikke · · Score: 3, Insightful

      If they move to an open source file system, iTunes for Windows could easily include a ZFS driver.

      and since apple has all the rights and source to HFS(+)(Journaled) they can just as easily write a windows driver for it as well.
      --
      Don't call me back. Give me a call back. Bye. So yeah. But bye our, well, but alright we are on a shirt this chill.
    7. Re:The were going to use Reiser by Anonymous Coward · · Score: 1, Informative

      (Correct me if I'm wrong, but some sort of ZFS driver is in the Linux kernel, and Sun is open sourcing Solaris.)


      ZFS Driver is being written to use Fuse - GPL and CDDL code can't be mixed due to GPL's restrictive nature. Sun opensourced Solaris almost two years ago. Everything is at opensolaris.org.

    8. Re:The were going to use Reiser by TheVelvetFlamebait · · Score: 2, Insightful

      If you don't like geeks discussing their experiences with technology, you could always, y'know, stop reading slashdot!

      --
      You know, there is a difference between trolling and pointing out the flaws in your reasoning. Just saying.
    9. Re:The were going to use Reiser by SuiteSisterMary · · Score: 5, Funny

      Am I missing any other notorious Slashdot cliche?

      I don't use Slashdot cliches, you insensitive clod!

      --
      Vintage computer games and RPG books available. Email me if you're interested.
    10. Re:The were going to use Reiser by Ant+P. · · Score: 1

      I don't think ZFS would be the right choice for an iPod. It seems more like it's designed for servers with CPU time to burn on validating all the checksums. Maybe they should've used UDF.

    11. Re:The were going to use Reiser by nine-times · · Score: 1

      If they move to an open source file system, iTunes for Windows could easily include a ZFS driver.

      I'm not sure it's so easy to include ZFS support in Windows-- at least, I don't see why it would be harder than including HFS+ support. There's already support for reading HFS+ in Linux, but Linux doesn't support ZFS.

    12. Re:The were going to use Reiser by tpjunkie · · Score: 2

      In soviet russia, slashdot cliches use you!

    13. Re:The were going to use Reiser by iminplaya · · Score: 1

      ZFS cannot be added to the linux kernel due to licensing issues...

      Aaah, leave it to licansing to force another kludge upon us. I can only dream how clean the code could be, and how fast the machine can go, if not for having to "license" around everything.

      --
      What?
    14. Re:The were going to use Reiser by A_Non_Moose · · Score: 1

      But they killed that project.


      Funny, but only two possible replies come to mind:

      Why? Did the drive's name have a space in it?

      or

      Are they married to that decision?

      (yeah, yeah, poor taste, short bus to hell and all that jazz...)
      --
      Have you read the moderator guidelines? Well, have you, PUNK? (and I want a Karma: Gnarly option)
    15. Re:The were going to use Reiser by benbean · · Score: 1

      All your cliches belong to us.

      --
      It's a Unix system - I know this.
    16. Re:The were going to use Reiser by Anonymous Coward · · Score: 0

      Im in ur cliche, maykin u lol

    17. Re:The were going to use Reiser by l3mr · · Score: 1

      You must be new here.

      --
      The world always seems brighter when you've just made something that wasn't there before. - Neil Gaiman
    18. Re:The were going to use Reiser by mrkitty · · Score: 1

      Cliches dead at age 55.

      --
      Believe me, if I started murdering people, there would be none of you left.
    19. Re:The were going to use Reiser by object88 · · Score: 1

      I don't use Slashdot cliches, you insensitive clod!

      Holy crap; I need to read Slashdot less often. I just laughed out loud at that.

    20. Re:The were going to use Reiser by Znork · · Score: 1

      "GPL and CDDL code can't be mixed due to GPL's restrictive nature."

      You mean, the CDDL's restrictive nature. The main contention issues appear to be the CDDL's choice of venue restriction, and patent conflict auto-termination issue. IE, restrictions beyond the GPL.

      As the GPL protects the freedom of the GPL protected derived code from the restrictions imposed by the CDDL, you can hardly say the GPL is the restrictive license...

    21. Re:The were going to use Reiser by veganboyjosh · · Score: 1

      Netcraft confirms it!

    22. Re:The were going to use Reiser by Anonymous Coward · · Score: 0

      I cannot find this word "licanse" in the dictionary. What are you talking about?

      To grant authoritative permission or license to. That definition seems to fit.

    23. Re:The were going to use Reiser by Liinux · · Score: 1

      No, I'm New Here! No, wait...

    24. Re:The were going to use Reiser by Anonymous Coward · · Score: 0

      If you don't like anonymous dicks griefing people, you could always, y'know, stop reading slashdot!

    25. Re:The were going to use Reiser by Sancho · · Score: 2, Interesting

      You mean, the CDDL's restrictive nature. Semantics. Each license restricts you from using the other. The GPL is generally more restrictive, but since you have to relicense under the GPL, further restrictions to modified software cannot be added. The CDDL has a restriction (regarding attribution) which must accompany any CDDL derivative work. Therefore, the licenses are incompatible. If the CDDL did not require attribution, or if the GPL allowed further restrictions to be placed upon derivative works, the licenses would be compatible (as I understand the issue, at least).

      Sun's patents prevent a compatible reimplementation of ZFS from being imported into the Linux kernel, so you won't see that, either. The most we'll get is a CDDL-licensed FUSE module. And that sucks.
    26. Re:The were going to use Reiser by gambino21 · · Score: 1

      I, for one, welcome our new cliche using overlords.

    27. Re:The were going to use Reiser by Fex303 · · Score: 2, Funny

      Netcraft confirms it!

      I'm getting this engraved on my tombstone...

    28. Re:The were going to use Reiser by aichpvee · · Score: 1

      So is OSX going to be case-sensitive by default now as it should have been from the start?

      --
      The Farewell Tour II
    29. Re:The were going to use Reiser by burndive · · Score: 1
      Oh, you can add it all you want, you just can't distribute the binaries.

      Which, as a user of Gentoo, is just fine with me.

      Once you get it working, could you send me your patches?

      --
      ...because "hacker" sounds way sexier than "code drone."
    30. Re:The were going to use Reiser by jeremyp · · Score: 1

      HFS+ is open source.

      --
      All I want is a secure system where it's easy to do anything I want. Is that too much to ask ~~ Randall Munroe
    31. Re:The were going to use Reiser by innerweb · · Score: 2, Funny

      Normally a License, but under a full moon, it mutates and attacks projects that are not corrupted.

      --
      Freud might say that Intelligent Design is religion's ID.
    32. Re:The were going to use Reiser by StarfishOne · · Score: 1

      And Netcraft confirms this!!one

    33. Re:The were going to use Reiser by Anonymous Coward · · Score: 0
      I can only dream how clean the code could be, and how fast the machine can go, if not for having to "license" around everything.

      Good point. And I can only dream how comfortable I could be, how much stuff I could have, if not for having to "pay" for things.

    34. Re:The were going to use Reiser by wootest · · Score: 1

      In Korea, only old people use that joke.

    35. Re:The were going to use Reiser by petermgreen · · Score: 1

      There's already support for reading HFS+ in Linux
      and for writing it.

      Unfrotunately the write support doesn't support journaling meaning you have to turn off the journal in OS-X first.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    36. Re:The were going to use Reiser by nine-times · · Score: 1

      Yeah, the reason i put "reading" was that, IIRC, there's support in the kernel for reading/writing but not creating HFS+ partitions. I didn't want to say "writing" and have someone make a big deal about that small inaccuracy.

    37. Re:The were going to use Reiser by JimDaGeek · · Score: 1

      OS X can be case sensitive, HFS+ has support for it, though it is not the default. I guess Apple went this method to make things easier for MS Windows users.

      --
      General, you are listening to a machine! Do the world a favor and don't act like one.
    38. Re:The were going to use Reiser by Anonymous Coward · · Score: 0

      For you comment, I deduce that you live in Soviet Russia.

      Am I right?

    39. Re:The were going to use Reiser by rthille · · Score: 1

      Apple owns HFS+, they could include a windows driver for HFS+ in iTunes if it made sense. ZFS is hugely complex (Val, now a linux kernel engineer who worked on it, gave a talk at our local LUG), and porting it would I imagine be much harder than something (HFS+) that in some form has existed since ~1985...

      --
      Awesome furniture, accessories and cabinetry in Santa Rosa, CA: http://humanity-home.com/
    40. Re:The were going to use Reiser by petermgreen · · Score: 1

      afaict creating filesystems is never handled by the kernel and theres no real reason for it to be.

      http://gentoo-wiki.com/HOWTO_hfsplus claims there are tools for creating hfs plus filesystems and an ebuild for installing them on gentoo.

      i don't think there is a way to turn off the journal from linux though other than reformating which may be a problem if you are working with an existing hfs plus volume and can't easilly get it onto an OS-X box for journal removal first.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    41. Re:The were going to use Reiser by larry+bagina · · Score: 1

      Apple could have included HFS read/write abilities in the Windows version of iTunes if they wanted to. FAT32 is more "open" (in that pretty much everything can read it) than ZFS ever will be. ZFS is great, but it's a poor choice for an iPod.

      --
      Do you even lift?

      These aren't the 'roids you're looking for.

    42. Re:The were going to use Reiser by Watts+Martin · · Score: 1

      Apple went that method (filename case-insensivity) to make things easier for Mac users, actually. HFS+ predates OS X, and it's a direct descendant of HFS from 1985. Remember, OS X may have been, in effect, NextStep 5, but it also had to be MacOS 10 -- a lot of Mac programs depended on HFS's "resource fork" functionality (basically, metadata).

      Personally, I'm not sure I see a terribly compelling advantage to either approach. I'm not bothered by case-sensitive file systems at all, having worked with BSD and Linux for years. But you know, I'm not sure I've ever been in a position where it was vitally important to have files named "FOOBAR", "foobar", "Foobar" and "FoObAr" in the same directory.

    43. Re:The were going to use Reiser by eclectic4 · · Score: 1

      "I've been asked time and again "why doesn't my iPod work with Windows?"

      I would tell them that you should format your iPod for Windows, unless you are absolutely sure you will only use your iPod on a Mac and format is as such by pluggin it into a Mac and clicking "yes, reformat it". The Mac can read and use iTunes just fine with a Windows formatted iPod. I use my iPod on my Windows machine for file transfer and listening to the music on it, and my Mac at home is the place I keep my music and have it synced. And you should never, ever, ever, ever, use an MP3 player as a "backup device for my important files".

      Ever.

      --

      "The greatest obstacle to discovery is not ignorance - it is the illusion of knowledge." - Daniel Boorstin
    44. Re:The were going to use Reiser by TheVelvetFlamebait · · Score: 1

      Touché.

      --
      You know, there is a difference between trolling and pointing out the flaws in your reasoning. Just saying.
    45. Re:The were going to use Reiser by aliquis · · Score: 1

      So, you want your MP3-player to work as an USB Mass Storage device so you bought an iPod? Makes sense...

      I can buy the file size trouble thought.

    46. Re:The were going to use Reiser by Anonymous Coward · · Score: 0

      fuck FUSE

    47. Re:The were going to use Reiser by chrome · · Score: 1

      Ooh, that was sharp!

    48. Re:The were going to use Reiser by aliquis · · Score: 1

      Thought it's called HFSX when it's case sensetive.

    49. Re:The were going to use Reiser by adrian727 · · Score: 1

      I, for one, welcome our new cliches speaking overlords.

    50. Re:The were going to use Reiser by FST777 · · Score: 1

      ME TOO!!!

      --
      Free beer is never free as in speech. Free speech is always free as in beer.
    51. Re:The were going to use Reiser by scott_karana · · Score: 1

      I've always wondered why people don't make a version of ZFS that is designed to be patched to the Linux kernel for non-distributed use. As I recall, you can do nearly bloody anything with either piece of software, just so long as it's not DISTRIBUTED.
      Production servers and powerusers both don't seem to mind patching Linux, so patching in ZFS should be no big deal. Right?
      Distributions could send out binary blobs (OSI approved, just pre-compiled) too!

    52. Re:The were going to use Reiser by NMerriam · · Score: 1

      So is OSX going to be case-sensitive by default now as it should have been from the start?


      Normal users are not case-sensitive, so systems targeting normal users will continue to be case insensitive (but preserving).
      --
      Recursive: Adj. See Recursive.
  3. As Homer would say... by dsginter · · Score: 2, Funny

    Mmmm... Boiled Oceans!

    --
    More
    1. Re: As Homer would say... by Dolda2000 · · Score: 1

      For those who don't get the "boiling oceans" part of this joke, it is a reference to a quote by one of the ZFS developers from Sun. I don't remember the exact words off the top of my head, but the message was that "since a 128-bit file system (such as ZFS) cannot be filled by earth-based storage, you can't reach the limits of ZFS without boiling the oceans". I can't say I'm perfectly convinced of its truth value, but I'll leave that aside. I thought the joke was rather funny, at least.

    2. Re: As Homer would say... by afidel · · Score: 2, Interesting

      Actually he proved it with some cool math. Basically there is a certain amount of entropy which must be overcome to initialize a bit of data, multiply that small amount of energy by the number of bits in a 128bit filesystem and you get enough energy to bring all of the earths oceans to a boil. It was one of the best examples for the scale of large numbers I've come across.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
  4. I'm giving odds... by Telephone+Sanitizer · · Score: 5, Informative

    Well, not in THIS forum. But elsewhere.

    5:1 that it's not the default root file system in Leopard.

    The first bootable release of ZFS (not "BUILD," but "RELEASE") isn't even due until the Fall.

    I'm not alone in this skepticism. See this Ars story, for example.
    http://arstechnica.com/journals/apple.ars/2007/06/ 06/sun-ceo-jonathan-schwartz-zfs-to-be-the-file-sy stem-in-leopard

    1. Re:I'm giving odds... by Vellmont · · Score: 2, Insightful


      5:1 that it's not the default root file system in Leopard.

      It would be foolish to make any new technology that touches so many other applications and parts of the OS the default when you don't have to. It's much smarter to make it an option and try to shake out any problems that arise. Then make it the default at a later date.

      --
      AccountKiller
    2. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      But if im not mistaken, Leopard isn't available until october anyway so that wouldn't be a problem.

    3. Re:I'm giving odds... by FuturePastNow · · Score: 4, Interesting

      The first bootable release of ZFS (not "BUILD," but "RELEASE") isn't even due until the Fall.

      OSX 10.5 ain't due 'til Fall, either.

      --
      Give a man fire, and you warm him for the night. Set a man on fire, and you warm him for the rest of his life.
    4. Re:I'm giving odds... by DurendalMac · · Score: 1

      5:1 that it is. You can bet your butt that Apple's working like hell to make it bootable. Now, take a look at Time Machine and tell me that doesn't sound a lot like ZFS.

    5. Re:I'm giving odds... by 0xABADC0DA · · Score: 5, Informative

      Are you kidding? This is ZFS we're talking about.

      ZFS is several orders of magnitude better at streaming large files like are used in video editing, which is already a huge draw for Macs. Since it is copy-on-write, writes are done without seeking so are very fast and can be spread out across multiple drives in parallel. IIRC within a zfs pool (collection of drives) you can make different 'filesystems' mirrored or striped, so you can have a /video that is striped and ultra-fast whereas /home is mirrored and fault-tolerant.

      You can take your 100gb video and instantly say 'snapshot this' then make any number of changes to it and if you don't like it just revert back again. Contrast to every other filesystem (besides spirolog) where you have to make a 100gb copy as a backup -- which takes forever, so nobody does it unless they have to.

      You can drop in a new drive and say 'use this drive' and your existing filesystem instantly has more space available and it is more fault tolerant or faster or both. If you want to remove a drive you say 'dont use this drive' and you can still use the OS normally while it moves data off to other drives.

      Something like ZFS, that "touches so many other applications and parts of the OS" has to be the default. Otherwise you have to support two completely different ways of using the system. And that bloat and complication costs a lot more than just getting it right through extensive testing. If you are really worried about it, don't upgrade the OS for a while.

    6. Re:I'm giving odds... by Vellmont · · Score: 2, Insightful


      Are you kidding? This is ZFS we're talking about.

      Right, I'm sure it won't wind up breaking some important application at the expense of adding a wiz-bang feature that 95% of the users couldn't care less about.

      I'm not a Mac user, but even if I could (maybe I can) add ZFS to my Linux workstation I wouldn't. I prefer stable and reliable over untested new features. I think most people feel the same way, so making the default something else makes a lot of sense. If ZFS is as great as you say, it will eventually become the default, and anything it breaks that's important will have been fixed.

      --
      AccountKiller
    7. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Opensolaris version of ZFS is Bootable.

    8. Re:I'm giving odds... by John+Allsup · · Score: 1

      Alternatively, Apple may choose to use a small HFS+ boot partition to get the system into a state where it can finish booting from ZFS.

      --
      John_Chalisque
    9. Re:I'm giving odds... by jellomizer · · Score: 5, Interesting

      If you were a Mac User you realize that Apple does stuff like this a lot, and they are quite good at it too.

      The Move from Classic (OS 9) to OS X forced people to Recompile/Port or Die from obsoleteness modernized almost all the software for Mac OS X. This removed a lot of Old Hacky code from the code base and forced developers to follow a more modern programming style.

      Next it was the move from Power PC to Intel. This once again required a full recompile but this time is assured that the recompile was with their own development tools. So more hacky code was removed and replaced with more standardized system calls.

      Now with ZFS on Mac OS X it is more likely that most things will work just fine with ZFS because Apple Knows what most of the calls to the OS will be. And the bulk of the legicy code has been updated.

      Windows, Linux and traditional Unix OS Devlopers don't normally Break Compatibility so often so their hacks to work around a shortfall in an OLD version of the OS holds threw to the following versions of their software on newer versions of the OS. So migrating OS ZFS on Linux is much more risky then moving to ZFS on OS X.

      But it is a trade off of getting Modern Software and paying more $$$ for the software. or Pay less for the software but make it hard to upgrade to a better system in the future.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    10. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      >You can take your 100gb video and instantly say 'snapshot this' then make any number of changes to it and if you don't like it just revert back again. Contrast to every other filesystem (besides spirolog) where you have to make a 100gb copy as a backup -- which takes forever, so nobody does it unless they have to.

      This isn't strictly true - essentially all linux filesystems support this snapshot feature through lvm (its done below the filesystem level).

    11. Re:I'm giving odds... by Guy+Harris · · Score: 1

      Something like ZFS, that "touches so many other applications and parts of the OS" has to be the default. Otherwise you have to support two completely different ways of using the system.

      Which certainly explains why Solaris 10 only supports ZFS, not UFS.

      Oh, wait, it doesn't....

    12. Re:I'm giving odds... by Anonymous Coward · · Score: 1, Informative

      zfs has been bootable on x86 Solaris using grub for over a month now zfs boot

    13. Re:I'm giving odds... by Anarke_Incarnate · · Score: 3, Interesting

      After 2 months of trying ZFS out in a non-prod environment, we reverted to UFS because ZFS was not as fast on 8k random reads and appeared to use a ton more RAM. In actuality, it used less than perceived but try explaining that to developers and DBAs. UFS will use system RAM for cache and report it as free RAM, then relase it when another process needs it. ZFS does similar things (with worse memory accounting) though if it uses system RAM for cache it reports it as used. A bug had to be squashed in ZFS regarding purging pages of cache when the system requested it back, because it would page out faster than it could account for doing so and cause massive thrashing in RAM. That has been fixed, however.

    14. Re:I'm giving odds... by Hatta · · Score: 1

      You can drop in a new drive and say 'use this drive' and your existing filesystem instantly has more space available and it is more fault tolerant or faster or both. If you want to remove a drive you say 'dont use this drive' and you can still use the OS normally while it moves data off to other drives.

      How exactly is this done? How do I tell ZFS whether I want mirrored or striped RAID? It can't be as simple as just letting ZFS figure it all out. I can't even find out how to tell ZFS how much redundancy to use.

      --
      Give me Classic Slashdot or give me death!
    15. Re:I'm giving odds... by BeProf · · Score: 1

      > You can take your 100gb video and instantly say 'snapshot this' then make any number of changes to it and if you don't like it just revert back again.

      Funny. That sounds almost exactly like Time Machine.

      --
      You are attempting to read sigs. Cancel or Allow?
    16. Re:I'm giving odds... by rho · · Score: 2, Interesting

      Apple has traditionally been in favor of forcing devices, such as not putting arrow keys on the original Mac keyboard.

      Their philosophy tends to be of a benevolent dictator like Linus. Apple is going to make you do some things, and it's for your own good. If you're not happy with it, usually you can do something else if you have the technical skill, or you can just go get a Winders box. This has meant Apple's been able to do things no other company could, and is also why, IMO, they're the top of the heap for consumer OSes.

      --
      Potato chips are a by-yourself food.
    17. Re:I'm giving odds... by suggsjc · · Score: 1

      I prefer stable and reliable over untested new features. I think most people feel the same way
      I think "most people" is an overstatement. "Most people" aren't tech minded geeks that have a clue about what the underlying technology is. I would imagine that if you could have a dialog that said something to the effect of:
      Which filesystem do you want to use?
      a) ZFS - Newest filesystem allowing for instant backups and versioning blah blah (unstable)
      b) Doesn't matter what the other choice is (stable)

      Out of this experiment 74% of the people wouldn't even read it and just choose the default value. If there wasn't a default value then of the original 74%, 58% of those would just choose the first one in the list, and the other 37% would take the newest wiz-bang features and 5% would have meant to have chosen ZFS but somehow still managed to click the wrong thing.
      Then out of the remaining 26% that actually read it only 37% would make an informed decision with about a 50/50 split for each, then 32% would choose the wiz-bang feature just because leaving 31% choosing the tested/stable version.

      All in all that is going to only leave roughly 11.76% (assuming ZFS wasn't the default choice) of people actually using a tested/stable version which unless I am mistaken does not qualify as "most people" and by the way 31% of those people actually meant to choose ZFS but just screwed it up.
      --
      When I have a kid, I want to put him in one of those strollers for twins and then run around the mall looking frantic.
    18. Re:I'm giving odds... by Znork · · Score: 1

      "Contrast to every other filesystem"

      Well, except every other filesystem on LVM as snapshots have been a feature of LVM since, like, forever...

      "If you are really worried about it, don't upgrade the OS for a while."

      Or implement the functionality as distinct and separate replaceable layers. Small parts, each good at their own thing, combined to form a flexible and powerful whole. Reminds me of the traditional philiosophy of that old OS, what was it called...

    19. Re:I'm giving odds... by piojo · · Score: 1

      zpool add my_existing_fs mirror /dev/new-device-1 /dev/new-device-2
      adds mirrored storage to an existing FS, while
      zpool add my_existing_fs raidz /dev/new-device-1 /dev/new-device-2 /dev/new-device-3
      adds raided storage. Or if you don't care about redundancy:
      zpool add my_existing_fs /dev/new-device-1

      --
      A cat can't teach a dog to bark.
    20. Re:I'm giving odds... by Lennie · · Score: 1

      If you don't know, why don't you just take a look:

      http://www.opensolaris.org/os/community/zfs/demos/ basics/

      --
      New things are always on the horizon
    21. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Apparently you Have a non-Backwards compatible keyboard Because Your shift Key seems To be Malfunctioning.

    22. Re:I'm giving odds... by Tuzanor · · Score: 3, Informative

      You are right to a certain extent, but you have to realized that current file systems are old and clunky. For a desktop or a few non-critical servers moving to the new tech is a great idea. Down the road, when ZFS is more mature and understood, it's going to be a welcome addition to most production setups. If you ran real-world mission-critical prod setups needing high availability you'd understand.

      Imaging you have a huge medical database on several servers and are running out of disk space. To expand, you need to plug in new hard drives, create RAID setups, create partions, move data over, restart the database, verify again and again, downtimes, etc. You can easily and efficiently grow file systems (unless you're using an expensive piece of software like veritas volume manager. With zfs, all I need to do to expand disk space in current WORKING filesytems is:

      zpool add oraclefs mirror c1t1d0 c2t1d0

      No luns to deal with. No other filesystem bullshit. You have no idea how excited this makes me for services that require large amounts of growing storage.

      Read up on zfs here: zfshttp://www.opensolaris.org/os/community/zfs/> It is the best thing to come out of Sun in a long time.

    23. Re:I'm giving odds... by Vellmont · · Score: 1

      I guess I don't understand what you're really getting at here. But in my experience if you ask someone "hey, do you want all your software to "just work", or do you want this new feature with a decent chance you'll lose some data, or some programs won't work" "most people" will choose the stability.

      It doesn't really matter if people are technical or not, if they read your little statement or not. What matters is knowing what most people desire and then providing that. The job of a software developer (in this case Apple) is to do just that so the user doesn't have to.

      --
      AccountKiller
    24. Re:I'm giving odds... by Vellmont · · Score: 1


      Imaging you have a huge medical database on several servers and are running out of disk space

      Heh. The biggest mistake in the world I could make was run my database on a brand new filesystem, very likely unsupported by the DB vendor. Running out of disk space would be the least of my problems. Losing data and DB crashed from an untested, brand new filesystem would be a MUCH greater concern than the very foreseeable and easy to solve problem as adding disk space.

      I've no doubt that ZFS or anything similar to it is a great solution. I just don't think it should be the default yet.

      --
      AccountKiller
    25. Re:I'm giving odds... by Sancho · · Score: 1

      I was definitely under the impression that you couldn't add to a raidz. Have you tried this, yourself?

    26. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      ZFS is several orders of magnitude better at streaming large files like are used in video editing


      Than what? HFS+? Prove it.

      (Don't bother, you can't, because you made a ridiculous claim. HFS+ does just as well as any other FS at this; it can stream reads and writes at or very close to the streaming performance limit of whatever I/O subsystem the HFS+ FS resides upon.)

      Since it is copy-on-write, writes are done without seeking so are very fast and can be spread out across multiple drives in parallel.


      What does copy-on-write have to do with seeking or spreading writes across many drives?

      Nothing, that's what. Please, stop being a typical ZFS fanboi and actually learn about filesystems.

      Something like ZFS, that "touches so many other applications and parts of the OS" has to be the default. Otherwise you have to support two completely different ways of using the system. And that bloat and complication costs a lot more than just getting it right through extensive testing.


      The whole point of a filesystem is that it abstracts the details of how information is stored on a storage medium. Application authors don't use different APIs for accessing different filesystems.

      Therefore, applications should not have to care whether ZFS is the underlying filesystem. And they won't need to, if Apple gets the MacOS X ZFS implementation right. There is absolutely no need for Apple to cause any short term problems just because some ignoramus fanboi thinks ZFS must go into everything Right Now. A long slow deployment will do just fine, thank you. There would not be any bloat or complication.

      (And it's not safe to assume Apple even intends ZFS to ever be the default FS. The vast majority of Apple's users don't need ZFS features, and Apple has a history of extending HFS+ to accomplish their goals.)
    27. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Regarding your signature, it's "more than once" not "more then once".

    28. Re:I'm giving odds... by piojo · · Score: 1

      Just tried it (with files, rather than real devices):

      paradox tmp # dd if=/dev/zero count=65 bs=1M of=/tmp/11 # for 6 files...
      65+0 records in
      65+0 records out
      68157440 bytes (68 MB) copied, 0.459183 s, 148 MB/s
      paradox tmp # zpool create test raidz /tmp/11 /tmp/22 /tmp/33
      paradox tmp # zpool add test raidz /tmp/55 /tmp/44 /tmp/66
      paradox tmp # zpool status test
      pool: test
      state: ONLINE
      scrub: none requested
      config:

      NAME STATE READ WRITE CKSUM
      test ONLINE 0 0 0
      raidz1 ONLINE 0 0 0
      /tmp/11 ONLINE 0 0 0
      /tmp/22 ONLINE 0 0 0
      /tmp/33 ONLINE 0 0 0
      raidz1 ONLINE 0 0 0
      /tmp/55 ONLINE 0 0 0
      /tmp/44 ONLINE 0 0 0
      /tmp/66 ONLINE 0 0 0

      errors: No known data errors
      --
      A cat can't teach a dog to bark.
    29. Re:I'm giving odds... by atrus · · Score: 1

      You can add multiple raidz vdevs to an existing pool. What you cannot do (for rather simple reasons) is add a raw device to a raidz vdev. The parity information would need to be moved around. This is where you dump, throw away the vdev, create a new larger one, and restore (which is what ZFS would have to do in the background anyway, but its hard and has bad data integrity problems to implement). Adding another raidz vdev to the pool will make ZFS dynamically stripe data over it. Or if you believe in overkill, you could even mirror your raidz vdevs...

    30. Re:I'm giving odds... by Electrum · · Score: 1

      Imaging you have a huge medical database on several servers and are running out of disk space.

      zpool add oraclefs mirror c1t1d0 c2t1d0


      With Oracle you can use ASM to manage the raw devices while the database is running. Most real databases including Oracle and PostgreSQL allow you to manage tablespaces online.

    31. Re:I'm giving odds... by divisionbyzero · · Score: 1

      Uh, you do know that Leopard was delayed until November-ish, right? Timelines seem to match to me.

    32. Re:I'm giving odds... by zeromemory · · Score: 2, Interesting

      IIRC within a zfs pool (collection of drives) you can make different 'filesystems' mirrored or striped, so you can have a /video that is striped and ultra-fast whereas /home is mirrored and fault-tolerant. No, you can't mix and match. When you create a zpool, you get to choose between the equivalents of RAID-0 (striped), RAID-1 (mirrored), and RAID-5/6 (striped with parity). You can nest the various RAID levels, but whatever you choose will apply to all the file systems in your storage pool. If you want to have /video and /home under different RAID levels, you'll need to create two different zpools. Now, this wouldn't be a big issue, as you could normally just partition up your drives to make the separate zpools, but Sun highly recommends allowing ZFS to manage entire disk drives (it's a volume manager along with a file system). If you create zpools out of partitions instead of drives, you lose a lot of neat features like automatic disk cache management and the soon-to-be-merged SMART awareness.
    33. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      You had me until you used the word "threw" instead of "through."

      Now that you are proficient in over 24 programming languages (according to your online resume), maybe you should take some time to learn English. Nobody is going to take your opinions seriously if you are writing at a 6th grade level.

      Also, your website and resume have a bunch of grammatical/spelling errors. I hope you don't point any prospective employers to your site.

      Peace.

    34. Re:I'm giving odds... by bfields · · Score: 1

      Something like ZFS, that "touches so many other applications and parts of the OS" has to be the default. Otherwise you have to support two completely different ways of using the system.

      Perhaps you haven't heard of the VFS? (A fine paper, by the way, well worth the read.)

      Supporting multiple filesystems is required in this case; users will not accept having to completely rebuild all their filesystems in order to use them under a new OS version, especially if doing so makes those filesystems inaccessible to earlier versions. "Default" in this case probably just means "default for newly created filesystems".

    35. Re:I'm giving odds... by k3vlar · · Score: 1
      Uhh. Actually Apple has been spamming by inbox nonstop about the fact that OS X 10.5 Beta will be available to people attending the WWDC.

      The Leopard beta. Available first at WWDC.
      The Apple Worldwide Developers Conference is less than a week away. Don't miss your chance to see a feature-complete version of Mac OS X Leopard, and take home a beta copy.
      --
      Unlike porn, which yada yada rimshot hey-ooh!
    36. Re:I'm giving odds... by suggsjc · · Score: 2, Insightful

      Ok, well we can agree to disagree but I would think that it have have very little to do with the question and everything with how your worded it.

      For example:
      If you asked do you want your software to "just work" or run the chance of having your computer burst into flames just to add some eye-candy/whatever? Then I would image that most people would opt for the stable approach.
      However, if you asked do you want to try out feature X? It will make your windows semi-transparent and ALL of your wildest dreams come true. It hasn't been fully tested, but it seems to be pretty stable. Then I would think you'd see a surprising number of semi-transparent windows.

      People really don't know and don't care about what happens behind the scenes. Also, I would think that most people would assume newer == better and that a higher version will be better, because hey...what can go wrong?

      --
      When I have a kid, I want to put him in one of those strollers for twins and then run around the mall looking frantic.
    37. Re:I'm giving odds... by TheNetAvenger · · Score: 2, Informative

      Contrast to every other filesystem (besides spirolog) where you have to make a 100gb copy as a backup -- which takes forever, so nobody does it unless they have to.


      Except of course NTFS which has been doing crap like this for several years now. So I'm not sure how you can seriously say 'every other filesystem'.

      This announcement isn't as exciting to the Windows world as it is to the OS X and *nix world where the features being offered by ZFS have not been available or consolidated into one FS model.

      NTFS on the other hand has been doing this stuff for quite some time, although ZFS does raise the bar even beyond NTFS, as it makes the jump from Terabytes to Exibytes or unlimited storage, even though there is not much contrast in 'features' offered by the FS beyond storage limits.

      I actually hope OS X does get ZFS in place and makes it default, as the time machine features would benefit from it greatly, as Windows 2003 and Vista use features of NTFS to make its 'Previous Versions' quite painless in terms of performance. On other FS that are not ZFS or NTFS, implementing a feature like this would be a serious performance impact.

    38. Re:I'm giving odds... by Vellmont · · Score: 1

      You're focusing too much on the question. The question is irrelevant and only serves to illustrate a point. I'm not doing a survey here, I'm trying to say what I believe "most people" want. "most people" just want to get whatever the task they're trying to accomplish done. Most of the time that means just doing what they've done before.

      --
      AccountKiller
    39. Re:I'm giving odds... by wootest · · Score: 1

      Time Machine is once-a-day backup of modified files. ZFS is ongoing versioning.

    40. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Could you not capitalise random words in your Rant please? It makes it Hard to read and it looks really Stupid.

    41. Re:I'm giving odds... by Anonymous Coward · · Score: 1, Funny

      Shit, you're absolutely correct. Since it's a beta surely nothing will change until it goes gold. Thanks for point that out to all of us.......

    42. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Greg,

      The reason for making ZFS the default is not for the POSIX layer. Who cares what read(2) does as long as it works. The point is for the GUI and users, do they get panels with lots of "if zfs then timemachine controls active" style mess. Does the system work by default and support features of zsh, or does the user only get features from decades past. Does the drive manager support both logical volumes and pools or just pools. If you think it should be both you obviously do not understand Apple's success.

      Copy on write has to do with seeking in that when you are writing data generally it all goes to where the head is. That's what it means. It also means that the data does not have to be allocated to other drives already in order to be written there (you don't have to run a block allocator to figure out where to put the data on another drive). Less seeking means more throughput.

      Nobody directly compares ZFS to Apple's HFS+ since HFS+ is not an interesting filesystem to benchmark. Compare reads while doing a lot of writes:

      Fs / unloaded / loaded
      UFS 0:50.2 5:50
      ZFS 0:31.8 0:36.0
      ext3 0:36.3 54:21
      reiserfs 0:33.4 69:45

      Unless you mean 'just as good at writing streams' to mean 'only if doing nothing else at the same time' then yeah, go ahead and use your FAT32 or whatever other old-school filesystem you want.

    43. Re:I'm giving odds... by Sancho · · Score: 1

      That makes a lot of sense.

      So if you stripe two raidz pools, you get some weird failure states.. One drive from each pool can fail, but two drives from one pool failing is fatal to the whole pool? Not quite raidz2.

      Presumably, also, the pools don't have to be of the same size, or contain similar numbers of hard drives. I'll have to break down the numbers on failure scenarios in this setup, but my gut tells me that it's not a great situation.

    44. Re:I'm giving odds... by 0xABADC0DA · · Score: 1

      "However, after I take a snapshot of a volume, the write performance to the original volume is
      degraded by about 5 times. Why such a big degrade?"

      "The advice seems to be to take the snapshot, mount it, copy the data off, and remove it as quickly as possible."

      If you are using LVM snapshots for making a Time Machine you might as well do your video editing in vmware and use their snapshots. Either way it's going to be slow as hell. ZFS snapshot and LVM snapshots are worlds apart.

    45. Re:I'm giving odds... by Sancho · · Score: 1

      So you end up with one filesystem consisting of two raidz pools. I don't think you actually gain redundancy in this scenario (as 0xABADC0DA implied). At best, you don't lose any, but I think you sacrifice a tiny bit.

      Statistically, before you add the second raidz, your filesystem has to have 66% of its drives fail before you lose your data. After adding the pool, if as little as 33% of your drives fail (2 out of the 6, if they're in the same raidz) you lose your data. In fact, if 50% of your drives fail, you've lost your data.

      With pools of 4 drives each, of course, the numbers get worse, but at least you get divisible percentages that you can realistically compare.
      1 raidz: 25% fail = ok
      1 raidz: 50% fail = data loss
      2 raidz: 25% fail = ok, if the two drives that fail are in different raidz
      2 raidz: 37.5% fail = data loss

      Interesting stuff, though.

    46. Re:I'm giving odds... by phatsphere · · Score: 1

      .. you forgot about the intelligent prefetching.

    47. Re:I'm giving odds... by Scaba · · Score: 3, Insightful

      So, how exactly does one roll back changes to a file on an NTFS partition?

    48. Re:I'm giving odds... by phatsphere · · Score: 1

      aside of that, there is also a "copies" parameter. default is 1, if set higher, all new files are stored this number of times - distributed across disks. this means, as far as i understand. two raidz with two disks and copies set to 2 saves everything in both raidz. redundancy is higher, not much space gained and speed is possibly higher, too. three raidz a two disks with copies 2 ... well .. do the math ;) [and there is a raidz2, too]

    49. Re:I'm giving odds... by dbIII · · Score: 2, Informative

      True - NTFS still has a major role in the enterprise of making NFS from a 10 year old Sun over 10mb/s look fast.

    50. Re:I'm giving odds... by Anarke_Incarnate · · Score: 1
      You forgot to set the blocksize to oracle's preferred 8K. You also forgot to set the read size to 8k so you don't take in 128K pages every time you only want 8K random reads. You also forgot to limit the ARC size (buffers) so that ORACLE will not have to contend with ZFS for memory resources.

      You fail it

    51. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      What does copy-on-write have to do with seeking or spreading writes across many drives?

      With a normal fs, writing over a file means seeking out the original blocks and writing data to them. With ZFS, new blocks are written to. With a normal fs, a file exists on one single drive. With ZFS, it can be split up over multiple drives. Writing 20 MB of data on 2 drives is faster than writing 40 MB of data on one drive. Jackass.

    52. Re:I'm giving odds... by 0xABADC0DA · · Score: 3, Interesting

      Really?

      NTFS has data checksums to detect and repair corruption caused by any component?
      You can add and remove disk space from an NTFS volume dynamically?
      NTFS does data-level journaling not to mention without the overhead of multiple writes of the data?
      NTFS can use compression without getting horrible fragmented or other negative side effects?
      NTFS snapshots do not affect performance of the normal system?
      NTFS has variable block sizes?
      NTFS is open source and took less than a decade to get support on multiple systems?

      As far as I know that's a big no on all those. I mean NTFS is very complex and has a lot of bullet points, but to claim that ZFS is just 'ntfs with larger address space' is really missing the boat.

    53. Re:I'm giving odds... by Sancho · · Score: 1

      It seems like the 'copies' parameter would only protect against disk corruption (sector dying, for example), rather than disk failure. I'll have to read up on what that actually does, because I thought ZFS was already supposed to protect data integrity pretty well.

    54. Re:I'm giving odds... by JimDaGeek · · Score: 4, Informative

      The Move from Classic (OS 9) to OS X forced people to Recompile/Port or Die from obsoleteness
      Not completely true. You can still run classic code, if you really want. I think what Apple did was make cocoa so much better that developers wanted it and users demand it.

      Next it was the move from Power PC to Intel.
      Again, you can run PPC under Intel via Rosetta. Though getting a native Intel build always performs better. Some say they don't notice any difference. I disagree. For example, a PPC build of Photoshop is much slower than a Universal build of the new Photoshop.
      So Apple does leave backwards compatibility stuff there, however they make the new stuff so much better, that developers and users want to get it ASAP.

      Now compare this to Microsoft. While .Net with C# is an improvement for developers in productivity, there is really no gain/difference for users. For example I had to port a legacy VB 6 app to C#/.Net. The end users didn't know anything different about the app from their point of view. Just switching to .Net didn't make the app inherit any default functionality. Contrast this to Cocoa where an app get spell checking via NSSpellChecker.
      --
      General, you are listening to a machine! Do the world a favor and don't act like one.
    55. Re:I'm giving odds... by Koutarou · · Score: 0

      The king of handling files for video editing was XFS/XLV on SGI iron. You could do guaranteed I/O throughput reservation from a filesystem/lv and the system would steal every bit of cpu/io possible to assure you meet the reservation requirements.

      It was great technology, but was killed in the marketplace by cheap fast commodity hardware being "fast enough" that it wasn't needed any more.

    56. Re:I'm giving odds... by Anonymous Coward · · Score: 1, Interesting

      NTFS in Vista answers yes to all those, except the open source / multiple systems one obviously.

      -Steve

    57. Re:I'm giving odds... by daBass · · Score: 1

      Haha, you Northern Hemisphere suckers always being hit by delays!

      At least I will get it in Spring while you are still twiddling your thumbs.

    58. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      There's a bunch of exposed APIs to let you do it. They've been around for a while in the form of the Volume Shadow Copy changes APIs, but I know they've greatly expanded support for Vista.

      I don't know if you can from the command line anymore - in the Vista betas (all the way through RC2) you could

      transaction /start
      netstat -an > testfile.txt
      transaction /rollback

      and testfile.txt wouldn't exist any more. I think its around in Window Server 2008, and hopefully returning to Vista clients in a later update.

    59. Re:I'm giving odds... by slashthedot · · Score: 2, Interesting

      A video of guys showing off features of ZFS with USB sticks is available at:
      http://youtube.com/watch?v=1zw8V8g5eT0
      There's an english translation available for that somewhere.
      ZFS can do a lot of things with ease that other file systems either can't, or it takes quite an effort.

    60. Re:I'm giving odds... by atrus · · Score: 1

      Yes, pretty much. Raidz2 is a type of element of the pool, not 2 raidzs. So you can lose at most two drives from each raidz2 component and be ok (the mean time between data loss is very good for double parity).

      Mirror is of course the best in terms of both speed and reliability. ZFS supports N-way mirrors (multiple 3 way mirrors has a very high MTDL)

      http://www.solarisinternals.com/wiki/index.php/ZFS _Best_Practices_Guide

    61. Re:I'm giving odds... by Phat_Tony · · Score: 3, Informative

      "PPC build of Photoshop is much slower than a Universal build of the new Photoshop"

      The old PPC builds of Photoshop are also much slower on PPC than the only universal version, CS3. They moved Photoshop from Codewarrior to Xcode between CS2 and CS3, and it's the most massive rewrite they've ever done. So you can't distinguish how much of the speed difference of CS3 over CS2 on Intel is due to it being Intel native, and how much is simply due to it being faster.

      --
      Can anyone tell me how to set my sig on Slashdot?
    62. Re:I'm giving odds... by adolf · · Score: 1

      NTFS has data checksums to detect and repair corruption caused by any component? (emphasis mine) No file system can repair all random data corruption, even ZFS. Among those that can always detect it and will generally fix it, however, is NTFS. (yawn)

      Other things in NTFS and ZFS have in common:

      Dynamic volumes. Efficient journaling. Useful, fast compression. Automatic periodic snapshots in 2k3 and Vista. Variable block sizes (oh, boy).

      Open source? Not in the slightest.

      But then, ZFS is so fucking open that it's license is incompatible with the Linux kernel and also, I expect, anything else with the GNU. So, as a Linux user, I don't count that as being too detrimental -- I can't use NTFS -or- ZFS.

    63. Re:I'm giving odds... by Anonymous Coward · · Score: 0


      ZFS is several orders of magnitude better at streaming large files like are used in video editing, which is already a huge draw for Macs. Since it is copy-on-write, writes are done without seeking so are very fast and can be spread out across multiple drives in parallel. IIRC within a zfs pool (collection of drives) you can make different 'filesystems' mirrored or striped, so you can have a /video that is striped and ultra-fast whereas /home is mirrored and fault-tolerant.


      Copy-on-write isn't a panacea. It leads to fragmentation problems. People won't be so quick to trumpet ZFS if they have to defrag :) Don't flame me because this doesn't fit with your opinion... go read the open solaris mailing lists.

    64. Re:I'm giving odds... by cibyr · · Score: 1

      Most people want features AND stability. However a lot of people *need* stability (and often don't realize how much they need it). Apple will only set ZFS as the default file system for OS X if they are damn sure people won't be losing data left and right.

      --
      It's not exactly rocket surgery.
    65. Re:I'm giving odds... by jhol13 · · Score: 1

      Actually you cannot add redundancy (fault tolerancy) to an existing zpool. You can add space to a stripe, though. And, at least right now, you cannot remove a disk from a stripe (although I have heard it would be available "soon").

      I would be really surprised if ZFS were default on Apple. It is, after all, the first incarnation of the file system in OSX. Even Sun does not have bootable ZFS (opensolaris just have bootable mirrors, not raidz).

      The advantages it gives (easy management, snapshots, crc's, speed, ...) are big, but there is no need to put it as default. There is no extra bloat as the old file systems must be supported anyway.

      I myself will be using ZFS on my next (home) file server, no doubt.

    66. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      NTFS has data checksums to detect and repair corruption caused by any component? (emphasis mine) No file system can repair all random data corruption, even ZFS. Among those that can always detect it and will generally fix it, however, is NTFS. (yawn) ZFS does checksums on all data. NTFS does not do this. So that's still a big N-O. Also, since it is copy-on-write for all data, even meta-data, even on a non-raid volume ZFS can repair damage by going back to an unused previous version that is still around. So if you are editing a text file and the hardware writes it incorrectly ZFS can revert it back to what was previously saved. NTFS does not do this and in many cases it does not even know the data was corrupted.

      Other things in NTFS and ZFS have in common:

      Dynamic volumes. Efficient journaling. Useful, fast compression. Automatic periodic snapshots in 2k3 and Vista. Variable block sizes (oh, boy). You seem to have a very basic and fundamental misunderstanding of how ZFS operates, and how that is very different from systems like NTFS. For example:

      ZFS is not journaling, it's log-based. NTFS writes to disk 'this is what Im going to do, does it, then writes that it completed'. ZFS writes to disk then says 'advance the current state to this point' NTFS transaction are meta-data only, ZFS guarantees are all data.

      NTFS compression works by compressing the file in 64k chunks, and when these change size it leads to severe fragmentation. It also maps the file offset to blocks in the MFT, so large files get many extra blocks allocated for these (these previously caused ridiculously bad fragmentation, now only in some cases is it that bad). NTFS has to decompress and then recompress the whole segment if it's changed. It's essentially useless except in some rare cases (but it makes people feel good, that their files are 'compressed'). ZFS compression uses variable-sized blocks so that there is almost no wasted space, and it can compress with better ratios than NTFS (for one thing the algorithm is pluggable). ZFS compression only compresses the data that is written. There's essentially no reason not to use compression with ZFS. ZFS can also compress the fs metadata; this is only not done to make analysis of the disk easier.

      Snapshots. NTFS makes a shadow copy, which is copy-on-write meaning that when you change something the first time since a snapshot it takes extra time to do so. ZFS does not make a copy for a snapshot ever, you get the one write for the 'current' file data as it is written.

      Block size. NTFS can vary the block size on a per-volume basis, from 512 bytes to 256k (or more?), but unless the block size is exactly 4k then compression cannot be used (NTFS cannot change the block size because it is bitmap allocated). ZFS block size can vary by individual file.

      Open source? Not in the slightest.

      But then, ZFS is so fucking open that it's license is incompatible with the Linux kernel and also, I expect, anything else with the GNU. So, as a Linux user, I don't count that as being too detrimental -- I can't use NTFS -or- ZFS. I would expect somebody with a low slashdot id not to be completely obtuse, so I am assuming you are so old that you're just a grumpy old man. The point of it being open-source, whether it can be directly inserted into linux or not, is that people can examine it and create a GPL version. They can look at published documentation and other people's public discussions to make their version. None of this is possible with NTFS, which is why linux is stuck with "cross your fingers" incomplete implementations even after 15 years (and Microsoft too for that matter).

      ZFS is simply much, much better than NTFS. It does more with less, and is better at it. And yes, there is a FUSE ZFS filesystem that already supports more of ZFS than most OSS drivers do for NTFS.
    67. Re:I'm giving odds... by Scaba · · Score: 1

      So, you really can't do it yet, unless you're a beta tester, but one day hope to? That's somewhat different than "has been doing crap like this for several years now," don't you think?

      C:\>transaction
      'transaction' is not recognized as an internal or external command,
      operable program or batch file.
    68. Re:I'm giving odds... by LKM · · Score: 1

      Otherwise you have to support two completely different ways of using the system.

      Apple already does support HFS+ (case sensitive and case insensitive) and UFS as bootable file systems.

    69. Re:I'm giving odds... by irgu · · Score: 1

      NTFS is open source and took less than a decade to get support on multiple systems?

      Yes, NTFS-3G

    70. Re:I'm giving odds... by NMerriam · · Score: 1

      It would be foolish to make any new technology that touches so many other applications and parts of the OS the default when you don't have to.


      But they do "have to" if features of ZFS are fundamental to major new features of the OS that they advertise (such as Time Machine).

      Having a Mac commercial where a user plugs in a hard drive and has the space automagically added to his existing "drive" is EXACTLY the sort of "it just works" feature that Apple has banked billions of dollars on. "Step one: plug in your new drive. Step two: there is no step two!"

      Apple has never been shy about saying "as of tomorrow, all of our systems will use this new technology, because it's just better. If you used the documented APIs, it should work, if not, then you're screwed. You'd better fix anything you did wrong or prepare to have your switchboard explode." Indeed, Apple has broken many many apps with every release of OS X and continually said that the APIs were still not finalized so companies shouldn't be outsmarting themselves using existing but undocumented features in hopes they would be supported in the future. If it wasn't official, it could change at any moment, and often DID.
      --
      Recursive: Adj. See Recursive.
    71. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      So, how exactly does one roll back changes to a file on an NTFS partition?

      Vista or XP variations?

      Vista supports full transactional abilities. XP supports shadow and mirror copies that keep the previous untouched version of the file alive through subsuquent changes, known as 'Previous Versions' in Windows 2003 & Vista. Vista uses parts of the transactional model to implement the same fetaures going beyond just shadowing.

      For an end user, right click on the Volume, Folder, or File and select previous versions, then roll the Volume, Folder, or File to any previous version. There is also System Restore (which again in Vista has changed to use the transactional features of NTFS), that allows a users to rollback the OS state to several previous dates without affecting user data. (Fixing everything from file corruption, to failed installation, and even removing rogue applications from the system.)

      Not sure why you would have to ask, this is easy stuff to look up. It also makes me wonder how many people on SlashDot actually still 'use' Windows in any serious capacity if such simple concepts seem 'fancy'.

    72. Re:I'm giving odds... by TheNetAvenger · · Score: 2, Informative

      Not only am I surprised you wouldn't have looked this up before asking such stupid questions, but even more surprised that people are dumb enought to mark it as informative.

      In reference to your questions:

      NTFS has data checksums to detect and repair corruption caused by any component?
      You can add and remove disk space from an NTFS volume dynamically?
      NTFS does data-level journaling not to mention without the overhead of multiple writes of the data?
      NTFS can use compression without getting horrible fragmented or other negative side effects?
      NTFS snapshots do not affect performance of the normal system?
      NTFS has variable block sizes?
      NTFS is open source and took less than a decade to get support on multiple systems?


      Yes & No - All FS models implement checksuming features. Although, no it is not to the same checksum level as you are going for, although it is far less impressive or important than you seem to think it is.

      Yes - Dynamic adding and removing has been with NTFS for a long time. Vista even adds a newbie interface for everything from partition resizing to the old school featuers of dynamic volume spanning, RAID, etc.

      Yes - Go look up the original NT journal features from 1991, and the expanded features used in Vista.

      Yes - Compression offers no more fragmentation than normal NTFS writes. This is insane.

      Yes - Snapshots, oh yes, have you not heard of 'previous versions' or System restore, they are built on the NTFS's various snapshot abilities.

      No - NTFS does not support variable block sizes beyond the intial selection when formatting the volume.

      Yes - No, not open source, but I don't see MS suing anyone using it. *wink. I also see it being used on OS X, Linux, and other OSes without much trouble. It also was developed over the course of 1990-1992, and even the current versions in use in Vista only slightly vary because of the robust and extensible model NTFS was built upon.

      For a 1992 FS(NTFS), that STILL is ahead of MOST other FS avaiable, with the exception of a few features you can pick and chooses, you are making are really stupid argument here.

      I NEVER said NTFS was still superior, but was rather making a point that MANY of the features that make ZFS so attractive are features that have been in NTFS for a LONG TIME.

      If I was arguing NTFS was superior, I would have done a smart reply like yours comparing ZFS to NTFS:

      Does ZFS support encryption?
      Does ZFS have minimal CPU usage on small file writes?
      Does ZFS compression support multi-threading?
      Does ZFS accurately report in-use disk space, or does it have problems because of the snapshots reliance?
      Does ZFS support a high compression ratio?
      Does ZFS support Quotas?
      Does ZFS support 'online' pool recofiguring?

      As you notice, NTFS still has 'features' even ZFS doesn't if you want to pick each of them to death.

      Again I will state ZFS is a good set of ideas and does move the FS concepts forward by moving more of the models into the FS set. It also is 128bit and allows for almost infinite storage.

      ZFS has a lot of good things, but that doesn't mean that NTFS is an old dog or hasn't already been doing some of these features, even if they are not implemented in the same storage pool metaphor.

      So once again, for the Mac world, ZFS is an awesome way to go if they can get the performance in line with their needs. However it is STILL just catching up with NTFS which is very feature rich and very solid and won't be hitting any walls for storage sizes in the next 10-15 years.

      Am I not allowed to believe both ZFS and NTFS are good technologies?

    73. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      Although my german is rusty, it was an intersting Video.

      Again, I think people have totally missed my original point.

      ZFS is 'finally' pushing new ideas that are giving 'feature rich' FS technologies like NTFS a run for the money.

      I have a great deal of respect for NTFS, considering most of what is great about it was developed back in 1990 and continues to work well with a lot of features and very low data loss rates while still giving good performance.

      I also think ZFS is great and will push MS and NTFS to move forward. Neither one is a clear winner yet, as NTFS still has features ZFS doesn't have and vice-versa. Also NTFS is a MS baby and ZFS is open and serves the rest of the market.

      What I objected to is the direction the threads were moving stating OS X would be superior to Windows because of ZFS, when the features they are touting have been in NTFS and Windows for a long time.

      I actually like both FS technologies, and how ZFS continues to rock.

      I apologize that my original post seemed trollish if that is how people took it.

    74. Re:I'm giving odds... by Tuzanor · · Score: 1
      That's all configurable with ZFS. with the blocksize option at creation time. much easier ufs on top of veritas.

      From the zfs docs:

      The ZFS adaptive replacement cache (ARC) tries to use most of a system's available memory to cache file system data. The default is to use all of physical memory except 1 Gbyte. As memory pressure increases, the ARC relinquishes memory.

      ZFS backs off when something else needs the memory. Still configurable if you want, though.

    75. Re:I'm giving odds... by Scaba · · Score: 1

      I guess I'm just too dumb to be using a computer. I changed some text in a file, and right-clicked on it in Explorer and didn't see any such menu item for "Previous Version."

      By the way - is your name Nick Burns?

    76. Re:I'm giving odds... by Lars+T. · · Score: 1

      Shit, you're absolutely correct. Since it's a beta surely nothing will change until it goes gold. Thanks for point that out to all of us....... Yeah, they are probably not going to fix minor bugs, but they will certainly replace the default file system.
      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

    77. Re:I'm giving odds... by 0xABADC0DA · · Score: 1

      NTFS has data checksums to detect and repair corruption caused by any component? Yes & No - All FS models implement checksuming features. Although, no it is not to the same checksum level as you are going for, although it is far less impressive or important than you seem to think it is. The answer is NO, NTFS does not do data checksums or any checksums at all. The only checksum it has is for the bootsector, and it doesn't check it and it is often wrong. What you seems to be talking about is meta-data only sanity checks. These are totally different from checksums. Ironically, NTFS does not have much problem with data corruption since drives are often unoptimally defragmented (a moved block gives the hardware a chance to detect weak sectors). ZFS does sanity checks and also checksums of all data and all metadata.

      NTFS does data-level journaling not to mention without the overhead of multiple writes of the data? Yes - Go look up the original NT journal features from 1991, and the expanded features used in Vista. NTFS does meta-data only journaling. It may be technically capable of doing data-level journaling (using multiple writes of the data) but it does not do this. ZFS always does the equivalent of data journaling and with no extra writes of the data.

      NTFS can use compression without getting horrible fragmented or other negative side effects? Yes - Compression offers no more fragmentation than normal NTFS writes. This is insane. NTFS compresses files in 64k segments, then writes these segments to the file packed one after another. When a segment grows larger due to data being modified, the file is fragmented. The location of compressed blocks within a file is stored in MFT linearly, and compressing a large file causes more MFT blocks to be allocated (often causing additional fragmentation). Changing one 4k block causes 64k to be decompressed then recompressed again. This is very, very inefficient compared to ZFS. For example, try using NTFS to compress a vmware image (do not try this on your actual system however...) and then use that system for a while. ZFS has no problems doing this.

      NTFS snapshots do not affect performance of the normal system? Yes - Snapshots, oh yes, have you not heard of 'previous versions' or System restore, they are built on the NTFS's various snapshot abilities. NTFS snapshots slow down writes to files. ZFS snapshots do not.

      For a 1992 FS(NTFS), that STILL is ahead of MOST other FS avaiable, with the exception of a few features you can pick and chooses, you are making are really stupid argument here. You said "NTFS on the other hand has been doing this stuff for quite some time" in response to features NTFS does not have. This is what I object to. Most NTFS features were done before in JFS, VxFS. NTFS does have more bullet points then most filesystems, but most filesystems implement each bullet point better than NTFS.

      So once again, for the Mac world, ZFS is an awesome way to go if they can get the performance in line with their needs. However it is STILL just catching up with NTFS which is very feature rich and very solid and won't be hitting any walls for storage sizes in the next 10-15 years No, you are still not getting it. ZFS has many features that NTFS does not have and never will because of its fundamental design. The reverse is also true to some extent, for example your point about ZFS and small writes/rewrites (although ZFS is already faster than UFS on this).

      Am I not allowed to believe both ZFS and NTFS are good technologies? If you believe that ZFS is merely 'catching up' to NTFS then you would be very wrong. You seem to believe that from the tone of your posts.
    78. Re:I'm giving odds... by mink · · Score: 1

      Vista alows RAID levels above 0? MS reps told me that I still have to buy a server version of windows if I want raid 10 or 5.
      Can you point me to some documentation for that?

      --
      Well I've wrestled with reality for thirty five years doctor, and I'm happy to say I finally won out over it.
    79. Re:I'm giving odds... by Anarke_Incarnate · · Score: 1
      Yes, I am aware of that. However, even with setting the ARC size in the kernel with parameter for a p size of 256MB (in hex) a c size of 256MB and a c max of 512MB it doesn't adhere to the limitations.

      http://www.solarisinternals.com/wiki/index.php/ZFS _Best_Practices_Guide#Memory_and_Dynamic_Reconfigu ration_Recommendations

      is a good place to start looking, but as you can see here:

      http://www.opensolaris.org/jive/thread.jspa?messag eID=122983

      You can use the command located here:
      http://milek.blogspot.com/2006/09/how-much-memory- does-zfs-consume.html

      to show how much is being consumed by ZIO buffers.

    80. Re:I'm giving odds... by Anarke_Incarnate · · Score: 1

      Also, forgot to mention we seem to be in agreement. I just meant that in the command given above, the 8K sizes were not set and that performance would suffer. I saw the same issue with Gemstone /S OODB.

    81. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      I guess I'm just too dumb to be using a computer. I changed some text in a file, and right-clicked on it in Explorer and didn't see any such menu item for "Previous Version."

      Maybe you are. I see 'Restore Previous Versions', that opens a list of versions dating back over a week.

      I can even do this on a Folder, and open the Folder how it looked at that time period without having to Restore anything, and just drag out items that were deleted or changed. It is kind of cool to view snapshots of Folder or Volumes from previous time periods.

      Vista also implements an extended function of this for photos in Photo Gallery, so all changes you make to the photo are preserved in case you want to 'undo' the changes at the FS level, and yet have immediate changes applied.

    82. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      These are totally different from checksums. Ironically, NTFS does not have much problem with data corruption since drives are often unoptimally defragmented (a moved block gives the hardware a chance to detect weak sectors). ZFS does sanity checks and also checksums of all data and all metadata.

      I won't even argue the checksum misinformation, but you need to realize that the journaling and log system of NTFS does use a form of checksuming, even though it is NOT the same type of checksums that people are talking about here.

      However I would like to redirect your attention to Windows 2008 Server, where NTFS is self healing, so even if there are hardware problems, the FS processes manage and correct any errors. (Oh and it uses checksums, just in case you are still hooked on this.)

      This is why I said 'Yes & No' with regard to checksums...

      No, you are still not getting it. ZFS has many features that NTFS does not have and never will because of its fundamental design. The reverse is also true to some extent, for example your point about ZFS and small writes/rewrites (although ZFS is already faster than UFS on this).

      How am I still not getting it. I never said NTFS would have these features. This is one reason I like ZFS because of the design model it uses that does make some of its features impossible in other FS paradigms. Quit trying to put words in my mouth, ok?

      If you believe that ZFS is merely 'catching up' to NTFS then you would be very wrong. You seem to believe that from the tone of your posts.

      I never said ZFS is catching up. I was referring to FS in general that have been trying to catch up, but have failed because they don't approach the FS model in a way that goes beyond the stuctures in most instances. NTFS is more than just the disk layout, as it requires process levels to maintain 'features', just like ZFS does.

      ZFS is newer than NTFS, and is newer to the game, and also has evolved just as NTFS has done. The first implementations of ZFS back in 1995 were not nearly as sweet as the current versions.

      Again, stop trying to put words in my mouth, or at least stop 'misunderstanding' me. Find someone else to talk down to, I am not the 'average' tech or SlashDot reader you can bloviate over.

    83. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      Vista alows RAID levels above 0? MS reps told me that I still have to buy a server version of windows if I want raid 10 or 5.
      Can you point me to some documentation for that?


      Vista being the 'workstation' level NT OS, only allows for RAID 0 as you state. I apologize if something I said led anyone to believe differently.

      If you are in need of RAID 5 or even RAID 1 Stripping, I would recommend a good 'removeable' controller, or common MB chipset that can easily be replaced in case of controller failure.

      Take Care.

    84. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      I won't even argue the checksum misinformation, but you need to realize that the journaling and log system of NTFS does use a form of checksuming, even though it is NOT the same type of checksums that people are talking about here. Show me a link. I have no idea what you are talking about when you redefine common computer science words like "checksum".

      However I would like to redirect your attention to Windows 2008 Server, where NTFS is self healing, so even if there are hardware problems, the FS processes manage and correct any errors. "Self healing" to Microsoft means it does a chkdisk in the background after a reset (like background fsck on FreeBSD). You may have been mislead by their terminology, but this is not anywhere close to what ZFS is doing.

      Also, it's 2007.

      Again, stop trying to put words in my mouth, or at least stop 'misunderstanding' me. Find someone else to talk down to, I am not the 'average' tech or SlashDot reader you can bloviate over. It's simple, if you want respect then get your facts straight and don't make insinuations unless you will stand behind them.
    85. Re:I'm giving odds... by slashthedot · · Score: 1

      Found the dubbed english version of the video. This would more interesting for those who don't know German:
      http://video.google.com/videoplay?docid=8100808442 979626078

    86. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      Show me a link. I have no idea what you are talking about when you redefine common computer science words like "checksum".

      What am I now your instructor? Go look up how log files in the journaling system of NTFS have always used checksums to ensure data writes. This is a 101 class topic, that you can find at Microsoft.com or even probably in Wiki. Many basic FS methods use a form of checksums for writes, let alone a FS as advanaced at NTFS that is journaling and logging these activities.

      "Self healing" to Microsoft means it does a chkdisk in the background after a reset (like background fsck on FreeBSD). You may have been mislead by their terminology, but this is not anywhere close to what ZFS is doing.

      Have you ever even used a Windows NT based OS? Your statement here calls this into question. NTFS runs a chkdsk only when the journaling checksums tell the system there is a need to run.

      This is why you won't see a chkdsk run on an NT based NTFS drive probably in the life of the computer, nor will a user initiated chkdsk find errors in the standard life of a NTFS volume on average unless there is hardware failure.

      You need to look up Windows 2008 Server, as is not a simple chkdsk nor something that is fired after a restart. It is a set of monitor processes that are ensuring integrity when there are hardware failures or block loss.

      It's simple, if you want respect then get your facts straight and don't make insinuations unless you will stand behind them.


      Stand behind facts? Is google or wikipedia too hard for you to use? I think you can footnote the information yourself, I am not here to 'teach' you. I will give you the facts, it is up to you if you want to learn and is is worth looking up for yourself; however, it is not my job to hand hold your research or curiosity.

      Once again I will state ZFS has a LOT of good ideas and does bring new ideas to the table, but this does not discount NTFS. Of all the Windows complaints and 'bashing', the NT Core and NTFS are not areas where you will find OS engineers finding much to poke fun at.

    87. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Show me a link. What am I now your instructor? Obviously not, since you've brought nothing useful to the conversation.

      Stand behind facts? Is google or wikipedia too hard for you to use? Funny because wikipedia also says you are wrong. See "Checksum/EEC". You've obviously been misinformed about NTFS on a number topics.
    88. Re:I'm giving odds... by TheNetAvenger · · Score: 1

      Funny because wikipedia also says you are wrong [wikipedia.org]. See "Checksum/EEC". You've obviously been misinformed about NTFS on a number topics.


      Well if wiki says it (in a different context none the less), then all the MS documentation going back to 1992 should be changed because we all know Wiki is superior to the company actually coding the product.

      Maintained Checksums as Wiki is referring and journaled or logged write data being checksumed are two different things.

      Again a freaking 101 class topic and people respond knowing less than nothing. And then have the nerve to bitch cause I didn't look up the information for everyone, sweet. (gag)

    89. Re:I'm giving odds... by Anonymous Coward · · Score: 0

      Google says your wrong, wikipedia says you are wrong, linux ntfs says you are wrong, ntfs-3g says you are wrong. But you respond that it's something everybody learns in their lol course using 15-year-old documents that apparently only exist inside Microsoft (since you can't post a link or quote them).

      You'll come back and argue on slashdot about something that even if it was true is irrelevant to the fact that NTFS doesn't do checksums on data or metadata, but you won't spend a minute pointing out which structure it is in from the link above to linux/fs/ntfs/logfile.h or correcting wikipedia. So it's not the time involved, or the desire to correct an alleged misconception, it's the fact that you look like a retard in this thread and are trying to save face. That's lame.

  5. Booting from ZFS? by rollthelosindice · · Score: 4, Interesting

    When ZFS was first mentioned in the same breath as OS X it was pointed out that at the time you couldn't boot off ZFS file systems, so people were thinking it would power external (or secondary) timemachine devices. If it's replacing everything, I'm assuming you can now boot from a ZFS drive? When was this functionality added?

    1. Re:Booting from ZFS? by Anonymous Coward · · Score: 0

      Yes, OpenSolaris has been able to boot from ZFS via GRUB for a while now.

    2. Re:Booting from ZFS? by AKAImBatman · · Score: 5, Informative

      When was this functionality added?

      March 28th, 2007 at 19 hundred and 50 hours Zulu time

    3. Re:Booting from ZFS? by Midnight+Thunder · · Score: 4, Interesting

      What would prevent you from being able to boot off a ZFS drive? Surely all that needs to be done is for Apple to add ZFS support to their EFI implementation?

      --
      Jumpstart the tartan drive.
    4. Re:Booting from ZFS? by Skapare · · Score: 4, Interesting

      I don't know any of the technology of ZFS, so I can only guess.

      For a boot loader like LILO, it will need to create a list of exact hardware datablocks to read the kernel in from. ZFS might move those blocks around after the "lilo" command built the block map. Then it can't load the kernel.

      For a boot loader like GRUB, it will need to have a read-only subset of the filesystem inside so it can find the kernel image file. That might be doable, but it hasn't been done, yet.

      So create a small boot partition on the first few megabytes of the drive, and make another partition for the rest and let it be a part of the ZFS pool (if ZFS can accept a partition, and not just a whole disk).

      A better option would be to get a computer that has legacy IDE support with bootability, in addition to the main SATA or SCSI support for major hard drives. Then add a Compact Flash adapter to the IDE port and use a small Compact Flash module to load the kernel from using your favorite boot loader. Or just use an all-SATA mainboard with a different Compact Flash adapter for SATA. A tiny CF memory module with 16MB or so would be enough to load a nice sized kernel. Or go with a 16GB one and have a copy of /opt and /usr on there as well (structured to work when mounted read-only).

      --
      now we need to go OSS in diesel cars
    5. Re:Booting from ZFS? by DECS · · Score: 2, Insightful

      Remember that Apple's Macs are EFI Intel PCs now. You don't need LILO and GRUB to start up an operating system, as EFI provides a minimal but sophisticated environment for handling multiple boot devices and system launching. It's like the Sun/Apple OpenFirmware that Macs have always had.

      You'd only need those things to get Mac OS X running on a DOS PC, or when using ZFS with Linux, right?

      ---
      Microsoft Surface: the Fine Clothes of a Naked Empire
      What happens when the core values of an empire are exposed as a fraud? Does it prompt it change? More likely, it results in the generation of more false information to cover up the embarrassing failings.

    6. Re:Booting from ZFS? by bahamat · · Score: 1

      What would prevent you from being able to boot off a ZFS drive?

      I thought the same thing for a long time. Afterall, it's just a filesystem, right? That isn't the case though. ZFS is a file system, logical volume manager, RAID system and more. Suffice it to say that booting from ZFS isn't trivial.

      But so what? What if, and I'm just speculating here, when you used DiskUtility to create a bootable Mac OS X partition it made a small HFS+ bootstrap partition for Mach that included all of the ZFS drivers before mounting the root? It would only need to be a few MB, so nobody would notice or even care if DiskUtility just silently created it on every disk. There are already plenty of things that Mac OS X does underneath its lickable interface that users never have to think about.

      Bootable schmootable. They can still make ZFS the default.
    7. Re:Booting from ZFS? by Anonymous Coward · · Score: 0

      I don't know about you, but I can see plenty of use for this for just data partitions. I never have any data that I would mind losing on my os partitions. With a ceiling of 48 Exabytes (1 with 27 0's), an infinite amount of drives can be used for one RAID volume.

    8. Re:Booting from ZFS? by adrianmonk · · Score: 1

      Or, given that you have the freedom to modify ZFS itself, since it has not yet been released as part of Mac OS (and compatibility with on-disk formats is not required), you could simply add support into ZFS for making a bootable region on the disk. That is, you could modify ZFS to have the option to cater to the needs of the bootloader. If that means making the boot image contiguous and start at some fixed offset from the beginning of the disk (or partition), ZFS can easily and safely rearrange data on the disk anyway, so it could do that if you wanted.

  6. He's already backpeddled by dancingmad · · Score: 5, Informative

    He's already taken it back, more or less:

    "I don't know Apple's product plans for Leopard so it certainly wouldn't be appropriate for me to confirm anything. [...] There certainly have been plenty of published reports from various sources that ZFS is in Leopard, I guess we will all have to wait until it is released to see if ZFS made it as the default, or if they simply announce that it will become the default in a future release."

    --
    "There is no time, sir, at which ties do not matter," Jeeves, (Jeeves and the Impending Doom)
    1. Re:He's already backpeddled by ArsonSmith · · Score: 1

      This seems to be a fairly large downfall in open source marketing. There really is no way to come out with a new software package and say "Surprise!! This is the best thing sense sliced bread!"

      If linux had similar plans for ZFS then it would be supported, improved upon, discussed the pros and cons of using it in open forum and eventual it'd just show up. No press release. No Touting of new features. It'd just work.

      In this way even a major update in opensource software is usually passed as, "Ohh, that's been working since the 0.09alpha3 patch 6 months ago. I thought this was news for nerd."

      --
      Paying taxes to buy civilization is like paying a hooker to buy love.
    2. Re:He's already backpeddled by WaZiX · · Score: 1

      Makes sense...

      I mean if time machine relied on ZFS, that would mean current users using HFS+ couldn't use it... or they'd have to somehow convert HFS+ filesystems into ZFS...

      Is that even possible without erasing everything on the disk?

    3. Re:He's already backpeddled by gclef · · Score: 5, Funny

      Translation:

      "I didn't know that Steve Jobs was going to call me and scream at me like a diseased monkey when I said that. I'm sorry Steve. Please don't kill me."

    4. Re:He's already backpeddled by Wesley+Felter · · Score: 1

      Some companies develop their open source projects initially in secret, exactly so that they can get a "big bang" of press when they release it. Novell seems to like this approach with XGL, the slab, etc.

    5. Re:He's already backpeddled by lskovlund · · Score: 0, Troll

      Methinks you've got some people mixed up here. That would be Ballmer, not Jobs.

    6. Re:He's already backpeddled by preggie_greggie · · Score: 1
      i think the following quote by ross shannon on that page is probably correct:

      My guess would be that Apple are going to advise/require users to format their dedicated Time Machine external drive as ZFS. Otherwise -- if it remains as HFS+ -- you get the situation where small changes to a file require an entirely new copy of it, which doesn't sound very useful at all. ZFS' snapshots are perfect, and don't require the boot drive to be similarly formatted.
    7. Re:He's already backpeddled by stefanlasiewski · · Score: 1

      There are three kinds of lies: Lies, Damn Lies, and Blog Comments.

      --
      "Can of worms? The can is open... the worms are everywhere."
    8. Re:He's already backpeddled by porl · · Score: 1

      you think this is a negative?

      the way i see it, 'surprising' people with new features might be great if you want the people at home to go 'ooh, look at this new fizdangle!', but people working seriously with it would prefer to know it has been tested in the real world and has stood up to all the initial reservations, criticisms and changes that most new technologies go through before throwing it into production.

      this is the reason why, for example, very few linux distributions have things like compiz/beryl enabled by default. some of them do, and that pleases the early adopters who like to see new 'surprises' without much fuss. most, however leave the 'sane' and well tested defaults so that people don't get the nasty surprises that often come with new surprises (take that, grammar nazis!).

      this is also why you have people talking about products like windows: "i'm not touching it till at least SP1 comes out" and so on: they might like some of the new features, but they aren't willing to risk the 'nasty surprises' and prefer to wait till it has been running in the real world to 'iron out' the problems for a while.

    9. Re:He's already backpeddled by ArsonSmith · · Score: 1

      You're trying to review the idea as an intelligent and logical person. Drink a six pack of Budwiser then reread my post. It may make more sense :) I didn't mean that the software was hurt by the process just the marketing. Marketing is usually done to the mindless masses to grab as much money as possible.

      --
      Paying taxes to buy civilization is like paying a hooker to buy love.
  7. It WAS... by WiseWeasel · · Score: 5, Funny

    Not anymore, it ain't... Now, Apple will go with NTFS just to spite them...

    --
    "I like systems, their application excepted", George Sand (French)
    1. Re:It WAS... by soft_guy · · Score: 1

      Do Windows users even use NTFS? My Win32 boxes are all FAT32.

      --
      Avoid Missing Ball for High Score
    2. Re:It WAS... by CthulhuDreamer · · Score: 3, Informative

      Microsoft crippled FAT32 in Windows 2000 and Windows XP. Both of these can only format a 32GB FAT32 drive, anything bigger requires NTFS (or a third-party formatter). FAT32 also has a 4GB file size limit which is an issue when dealing with large avi files and DVD rips.

    3. Re:It WAS... by drinkypoo · · Score: 1

      Do Windows users even use NTFS? My Win32 boxes are all FAT32.

      Thanks to the ntfs-3g project, both Linux and OSX can use FUSE to get full read/write support of NTFS volumes.

      Thus, yes, even my removable drives use NTFS at the moment.

      That will all change soon, and I will format them HFS+, because I am now running Windows only in virtual machines and on a system which exists solely to run video signage software, and Linux has HFS+ support as well - without journaling, but I use the disks in question for backup and temp storage only. This is all at work; at home, we still have NT machines, so I still use NTFS.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    4. Re:It WAS... by Anonymous Coward · · Score: 1, Funny

      Naw, Apple's really going to spite MS by implementing and using WinFS.

    5. Re:It WAS... by Vancorps · · Score: 2, Informative

      In Windows 2000 and XP you can format Fat32 up to 160gigs assuming you have the correct driver. With SP2 for XP you can format it up to 250gigs I believe. Most removable drives from Maxtor for instance were and are formatted Fat32.

      Technically you're right though, since most Linux distros can format fat32 up to 2tb. NTFS is vastly superior though so the issue has never really affected me personally. Of course on a Windows machines you don't have to use Microsoft's formatting or partitioning tools, you can always format fat32 up to 2tb on your linux box then put the disk into a Windows box and it will read it just fine. I can't imagine why you would want to do that but the option exists.

    6. Re:It WAS... by Anonymous Coward · · Score: 1, Funny

      Maybe they should use WinFS, so that MS has somewhere to steal it from

    7. Re:It WAS... by Anonymous Coward · · Score: 0

      I do that with my USB hard drives. I need them compatible with everything, including linux boxes too old to have useful NTFS support and Windows 98 machines. Works like a charm, never had any problems (the filesize limit did bite me once but I just added an ext3 partition alongside the FAT32).

    8. Re:It WAS... by __aamnbm3774 · · Score: 0

      sure, but your sector sizes would be enormous and you'd waste more space than Rosie O'Donnell.

    9. Re:It WAS... by Blakey+Rat · · Score: 1

      Every computer sold with Windows 2000 or Windows XP was sold with NTFS on it. That's pretty much every computer sold PERIOD after 2001. I doubt any Win32/NTFS computers have been sold since XP SP1 came out.

      Of course that's because nobody runs Win32 anymore. WTF? Upgrade from Windows ME, dude, you'll enjoy joining the rest of us in the 21st century.

    10. Re:It WAS... by bazorg · · Score: 1
      Of course on a Windows machines you don't have to use Microsoft's formatting or partitioning tools, you can always format fat32 up to 2tb on your linux box then put the disk into a Windows box and it will read it just fine. I can't imagine why you would want to do that but the option exists.

      If you visit Macrumors.com and its forums you'll find dozens of help requests everyday from people who bought a BIG external hard drive and need to format it in a way that will be R/W from both their Mac and their Windows PC. Nobody seems to care much about using EXT2 and the free drivers available for both MS Windows and OS X to use this partition standard, on the contrary, people are really keen on finding how to have their Macs use NTFS or FAT32 on 500GB disks.

    11. Re:It WAS... by Anonymous Coward · · Score: 0

      sure, but your sector sizes would be enormous and you'd waste more space than Rosie O'Donnell.

      Aw come on mod this guy up that was awfully damn funny.

    12. Re:It WAS... by incer · · Score: 1

      Or you can just insert your GParted live CD, reboot, format and boot in Windows again without having to touch anything inside the computer. It even works on my Macbook.

    13. Re:It WAS... by HaTaX · · Score: 1

      Win32 doesn't support NTFS anyway, so your only choice is to rock on with that Fat32 FS... Honestly though, NTFS is not that great and I will be glad to be able to goto a new FS.. Only thing I really wonder about is, will XP get a port of it? Cause I'm just not quite up for the whole vista thing, even XP has too much bloat for me... I've considered switching desktop environments just to get rid of explorer.. wow, that was a ramble..

    14. Re:It WAS... by Vancorps · · Score: 1

      I ran into the same problem with a photographer giving me photos. He's on a Mac and I'm on Windows. I booted into Knoppix, formatted the drive fat32 for him and then we were both happy.

    15. Re:It WAS... by Blakey+Rat · · Score: 1

      Actually, Windows XP, through some kind of insane hackiness, will actually boot and run from Fat32. Highly NOT recommended, though, I'm sure performance goes into the crapper.

    16. Re:It WAS... by iphayd · · Score: 1

      It's common in shops that use external hard drives on Macs and Windows to format them as large fat32. That way both OSes can read and write to them, since NTFS is read only on Mac OS X.

  8. No no no by Guanine · · Score: 5, Informative
    Then he retracted his statement, saying he didn't know if it was the _default_ or not. Here's his quote, from a link on Daring Fireball:

    I don't know Apple's product plans for Leopard so it certainly wouldn't be appropriate for me to confirm anything. [...] There certainly have been plenty of published reports from various sources that ZFS is in Leopard, I guess we will all have to wait until it is released to see if ZFS made it as the default, or if they simply announce that it will become the default in a future release.


    1. Re:No no no by Anonymous Coward · · Score: 0

      Well it looks like other developers already have pointed out. In the April 18th release you cannot install the base os on zfs. You can however create zfs volumes after you are installed. Much like the current versions of solaris 10 after 06/06.

      Finder chokes on stuff you make using zpool create/etc... however. Or at least it did the last time I tried.

      Lets face facts, zfs is in leopard betas, it looks like he hasn't seen the developer builds, or he has seen ones that the developer community hasn't. Finally, we don't know what apple plans to do yet for the real leopard. WWDC next week should help clarify this.

      Could it happen? Sure, but remember HFS is case preserving, moving off of that assumption with zfs (this is implementation dependent, apples zfs implementation could keep this behavior) will probably guarantee to break existing apps just like ufs does. Have you tried to run some older osx apps off of ufs? It can be ugly.

      I am guessing we will see zfs in 10.6 as a bootable option. 10.5 seems too soon considering opensolaris is the only os that current can boot zfs natively, soon solaris 10 ga will as well. Unless apple plays tricks like I used to on my home file server and do a tiny ufs/hfs filesystem bootstrap trickery.

      Lets wait until the production release in October. At this point even with more knowledge than the average slashdotter (I paid for it :), I don't think we should even bother speculating, even if I were a ceo of a company.

      Not just that but apple could always strip out the zfs support if they wanted. They haven't even said they will use it after all. Could be a harbinger of 10.6 that will be removed.

      Posting anonymously just in case...

  9. Time to go back to bed. by dancingmad · · Score: 0

    On my first read, I was wondered why Mark Hamil was making announcements about OS X file systems...

    --
    "There is no time, sir, at which ties do not matter," Jeeves, (Jeeves and the Impending Doom)
  10. What Makes Me Nervous by asphaltjesus · · Score: 1

    I have a particular gift for breaking things that are supposed to be reliable. (Admittedly, mostly through ignorance, but I digress)

    Besides the worrisome concepts of delayed writing and an always consistent file system, I can imagine never being able to bring a zfs back pretty easily. Which, the snapshots are supposed to solve, but pretty soon, my hardware storage budget just went through the roof because I'm storing terabytes of snapshots pretty quickly.

    --
    Got Trader Joe's? friendwich.com RSS feeds work now!
    1. Re:What Makes Me Nervous by DurendalMac · · Score: 1

      I believe you have the option to make snapshots of your system differential if you can, so you'll only be storing the changed data after the initial backup. That should save you a little space.

    2. Re:What Makes Me Nervous by PCM2 · · Score: 1

      pretty soon, my hardware storage budget just went through the roof because I'm storing terabytes of snapshots pretty quickly.

      Methinks you exaggerate a bit, but still ... have you looked into the cost of a terabyte lately?

      --
      Breakfast served all day!
    3. Re:What Makes Me Nervous by asphaltjesus · · Score: 1

      Get back to me when you price a terabyte of production quality fiber channel drives.

      I'm not talking about some sata drives for my desktop.

      --
      Got Trader Joe's? friendwich.com RSS feeds work now!
    4. Re:What Makes Me Nervous by PCM2 · · Score: 1

      You probably wouldn't be talking about a Mac, then, either.

      --
      Breakfast served all day!
    5. Re:What Makes Me Nervous by unDees · · Score: 1

      Please come to work at my company. As a developer, I love most of all the testers who have a knack for breaking my stuff. I don't always realize it at the time, but they're my favorites.

      --
      "I call a baby goat a 'goatse.'" -- my non-Internet-savvy 6-year-old stepdaughter
  11. Sounds good but.. by jshriverWVU · · Score: 1

    does the user get a choice? Having access to more filesystems is a plus, but I wouldnt want to replace HFS+ for ZFS personally. So hopefully you can choice which one you want. Similiar to how Windows lets you choice FAT32 and NTFS, or OS X you can do HFS or HFS+, or Linux with ext2, ext3, reiser, etc.

    1. Re:Sounds good but.. by lexarius · · Score: 1

      In the past, there has been a button in the installer that lets you specify somewhat more precisely how to do the installation. This includes a drop box that allows the user to choose among the (supported) filesystems included in that release. This isn't nearly as complex or feature-rich as is included in various Linux distribution installers, but is at least good enough for most power users. Anyone who needs more advanced features than this has probably noticed that before beginning the installer there is a menu which includes various utilities, one of which is the Disk Utility for all of your partitioning and RAIDing needs.

    2. Re:Sounds good but.. by Sancho · · Score: 1

      I'd rather a standardized filesystem be used. Otherwise, you're asking for compatibility problems (see the current issues with case-sensitivity if you choose to make your filesystem case-sensitive). In this case, I'd rather there be less choice for the main partition.

  12. I doubt it by ceswiedler · · Score: 4, Insightful

    Jobs is probably not happy about his thunder being stolen right before for the June 11th keynote

    I strongly doubt he didn't know about it. This is Jonathan Schwartz, not a OS X rumors blogger. At any rate, ZFS in OS X is Sun's thunder; Time Machine is Apple's thunder, and that's already announced. How many OS X users (other than slashdot readers) will care in the slightest about the underlying filesystem? What they care about are the features, like Time Machine, that it enables.

    1. Re:I doubt it by WhiteWolf666 · · Score: 2, Interesting

      Although Apple hates preemptive disclosure, this goes right along with their "OS X is industrial grade" strategy.

      All over the place Apple advertises that OS X is "Industrial UNIX at the core".

      Now, with ZFS, Apple can advertise having a next-generation omega filesystem to replace the long-in-the-tooth Journaled HFS+, which was significantly better than NTFS.

      NTFS versus ZFS is a joke ;-)

      --
      WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
    2. Re:I doubt it by dabraun · · Score: 1

      Now, with ZFS, Apple can advertise having a next-generation omega filesystem to replace the long-in-the-tooth Journaled HFS+, which was significantly better than NTFS.

      NTFS versus ZFS is a joke ;-)


      Care to back that up? I don't know the pros and cons of HFS, but it sounds like if they needed to change filesystems to support what is basically the same idea as Windows's system restore feature (time machine) then HFS wasn't "significantly better than NTFS. As to ZFS, can you elaborate on what makes it any better than NTFS (I'm not saying it isn't...)
    3. Re:I doubt it by ArsonSmith · · Score: 1

      "Z" is a way better letter than even both "N" and "T" put together.

      --
      Paying taxes to buy civilization is like paying a hooker to buy love.
    4. Re:I doubt it by SuiteSisterMary · · Score: 1

      Actually, I believe that what they're talking about is closer to NTFS's Volume Copy Shadow Service.

      NTFS isn't a bad file system, by any means; however, it does things differently than UNIX, and therefore, to some, is inherently flawed. Of course, these same people then turn around and talk about how bad a monoculture is...

      --
      Vintage computer games and RPG books available. Email me if you're interested.
    5. Re:I doubt it by dabraun · · Score: 1

      Actually, I believe that what they're talking about is closer to NTFS's Volume Copy Shadow Service.


      We are both right. System Restore is built on top of the Volume Shadow Copy Service. System restore is a user-visible feature (pervious versions tab on files in Vista, and the actual system restore checkpoint UI in both Vista and XP), Volume Shadow Copy is a technology used for this and other things (live backups of large server systems for example).
    6. Re:I doubt it by neuroklinik · · Score: 1

      Time Machine doesn't depend on ZFS, and works fine with HFS+.

      From Ars Technica:

      http://arstechnica.com/staff/fatbits.ars/2006/8/15 /4995

      "Time Machine does not use ZFS. There was a lot of initial confusion about this, partially because Leopard does include a port of DTrace, the "other" high-profile open source project to come out of Sun's OpenSolaris efforts. But the absence of ZFS was no surprise to me."

    7. Re:I doubt it by BosstonesOwn · · Score: 3, Insightful

      Also to add to your point.

      Their new 10.5 servers will probably be closely related to some products in the Sun stable at the moment. With a successful launch this could help start reversing the trend of server dominance.

      If mac starts adding more aspects of unix to the OS and gives it a pretty interface , windows small business server markets could be in trouble. I would personally like to see more of these boxes around.

      --
      This package Does Not Contain a Winner
    8. Re:I doubt it by Anonymous Coward · · Score: 1, Informative

      That's easy, where to begin?

      ZFS is better than ntfs because of:
      Transactional operation (always consistent on disk, even in a power outage)
      Checksums(and ability to recover data in flight using RAID info)
      Dynamic striping
      Dynamic file block sizes
      Limitless size and quantity of pools/FS/snapshots
      Automatic parallelism

      Should I keep going?

    9. Re:I doubt it by peacefinder · · Score: 1

      "windows small business server markets could be in trouble"

      It might be trouble for Microsoft, but for those of us in the target market it's delightful. :-)

      --
      With reasonable men I will reason; with humane men I will plead; but to tyrants I will give no quarter. -- William Lloyd
    10. Re:I doubt it by Hatta · · Score: 1

      Although Apple hates preemptive disclosure, this goes right along with their "OS X is industrial grade" strategy.

      If Apple wants to be taken seriously as "industrial grade" they need a real unix filesystem hierarchy.

      --
      Give me Classic Slashdot or give me death!
    11. Re:I doubt it by SuiteSisterMary · · Score: 1

      Maybe. But System Restore can be done without any sort of filesystem trickery; it's just making a copy of important files before you clobber them. Of course, a simple way of doing this, if you have the file system, is copy-on-write...

      Remember, System Restore was introduced with Windows ME.

      --
      Vintage computer games and RPG books available. Email me if you're interested.
    12. Re:I doubt it by Overly+Critical+Guy · · Score: 1

      The standard UNIX filesystem hierarchy FUCKING BLOWS. That said, Leopard will be UNIX certified.

      --
      "Sufferin' succotash."
    13. Re:I doubt it by Phroggy · · Score: 1

      How many OS X users (other than slashdot readers) will care in the slightest about the underlying filesystem? I'd imagine a significant number of the attendees at next week's WorldWide DEVELOPER Conference would be interested in the filesystem, and that's where Jobs would be making any surprise announcements about it.
      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
  13. Oh, great: another DiskWarrior lag by dpbsmith · · Score: 2

    Yes I'm sure it will be worth it in the long run but I'm not looking forward to yet another hiatus in which: no industrial-strength disk-recovery tools are available, in which accidentally running the wrong disk-repair tool on the wrong partition hoses it instead of fixing it, and in which yet more legacy software suffers breakage due to subtle incompatibilities in implementation.

    (Yesyesyes, I know, ZFS is reliable that disk-recovery tools are not needed. And if you believe that, then you probably believed Microsoft when they said NTFS volumes never needed defragmentation).

    Dear Apple:

    Please let HFS+ still be an option.

    Please let Classic still run on Power Mac processors.

    Please let reasonably well-behaved software that uses resource forks still work.

    Please let it be case-insensitive and case-preserving.

    1. Re:Oh, great: another DiskWarrior lag by drinkypoo · · Score: 2, Informative

      (Yesyesyes, I know, ZFS is reliable that disk-recovery tools are not needed. And if you believe that, then you probably believed Microsoft when they said NTFS volumes never needed defragmentation).

      Well, it's true! NTFS volumes never need defragmentation! On the other hand, Microsoft provides you with a defragmenter service (at least in 2k and later) and allows you to defragment files on NTFS volumes... :D

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    2. Re:Oh, great: another DiskWarrior lag by mini+me · · Score: 1

      then you probably believed Microsoft when they said NTFS volumes never needed defragmentation

      I don't have a Windows computer so I cannot back up the claims, but from what I've heard there is no real performance benefits from defragmenting NTFS as opposed to FAT where there is a significant improvement.
    3. Re:Oh, great: another DiskWarrior lag by Vancorps · · Score: 2, Insightful

      You also don't "need" to defragment the files. The FS is perfectly happy filling in the gaps with additional files. Performance will suffer but it will indeed work reliably.

    4. Re:Oh, great: another DiskWarrior lag by Guy+Harris · · Score: 2, Interesting

      Please let reasonably well-behaved software that uses resource forks still work.

      You are aware that ZFS - and, for that matter, Solaris's UFS - supports an arbitrary number of named forks in files? (Sun calls them "extended attributes", probably because that's what NFSv4 calls them, but they're really named forks/named streams/whatever you want to call them.)

    5. Re:Oh, great: another DiskWarrior lag by Wesley+Felter · · Score: 2, Informative

      Yesyesyes, I know, ZFS is reliable that disk-recovery tools are not needed.

      A common misconception. The "zfs scrub" command will scan the filesystem and try to correct any errors that are found (or panic the kernel); the difference is that ZFS can do this while the filesystem is mounted.

    6. Re:Oh, great: another DiskWarrior lag by frogstar_robot · · Score: 1

      You've heard wrong. Windows disks can be slowed down quite badly by fragmentation. I've cut long log-in times anywhere from one-half to two-thirds by defragging Windows clients and servers.

    7. Re:Oh, great: another DiskWarrior lag by drinkypoo · · Score: 1

      Actually, the biggest problem is fragmentation of system files, to wit the paging file, and the registry files. There's a tool called PageDefrag that you can configure to defrag all of these at boot every time; to make this not kick your ass, set the paging file to a fixed size.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    8. Re:Oh, great: another DiskWarrior lag by SeaFox · · Score: 1

      Please let it be case-insensitive and case-preserving.

      psst!
  14. Is that all? by o-hayo · · Score: 2, Interesting

    Maybe some in the know (not me) could fill us (people like me) in... Are there other benefits that will come from moving to ZFS? I'd guess that for the average consumer any performance gain, or loss, won't really make a difference, but what about those running servers or doing heavy video/audio work? Or are there other aspects of this filesystem that will make it that much better than HFS+?

    1. Re:Is that all? by kildurin · · Score: 3, Informative

      A couple. 1) ZFS does not require a format. It takes me around 2 seconds to create a pool on a raid disk. Also true on a 500G hard drive. 2) ZFS pools (or directories/partitions if you will) can span multiple drives. 3) ZFS pools can have volumes (or drives) added to them at will. This means that if I run out of space in my music folder, I can add storage to it by adding another drive and adding it to the pool where my music folder is. Hope that helps

    2. Re:Is that all? by tempest69 · · Score: 1
      The big issue is that it can get huge. They designed the filesystem so that it can handle an obscene amount of drive space and filesizes that make my mind feel a bit odd to comprehend. It is designed to hold up to moores law for another thirty years. Most Everything is 128 bit addressing. To put 128 bit addressing into some sort of perspective you could digitize the positions of all the protons/neutrons/electrons in your body to a unreasonably tight tolerance 10^-9 ang with vector data included, pump that much information down a wire every second, and not max out the filesystem in your lifetime.


      The issue that needs to be addressed is small files. The problem is that there are more and more small files in a system.. This becomes less efficient as drived attain higher throughput, as seek times are still kinda slow. assume a seek time of 4 ms, in order to read a file, the head must seek the filesystem ( a few times), find the file offset and seek the file start. so with just seek time youve used 8-32 ms to get a file. A one MB file takes 16ms to transfer on a common drive (Maxtor 320 SataII $79 60MB/s) so 500K files spend half of the time seeking at a minimum.


      However there are huge chunks of files that are 0-150k on a drive these still take the 8ms to access. Copying 50 thousand of these small files across a network is a monster, tarring them is a monster, even though you might only have a couple gigs of data. Reiser4 had a cute solution of putting small files directly into the file-system to reduce the seek time. My cute solution was to use a usb stick to hold the small files. But most file systems dont have a good solution yet. So small files become the inefficient moster of the server room.


      Storm

    3. Re:Is that all? by ms139us · · Score: 1

      That's not entirely true. ZFS pools consist of one or more concatenated vdevs, each of which can be raw partitions, RAID1, RAID5 or RAID6. ZFS gets grumpy if your pool consists of different types of vdevs, e.g. a raw disk vdev and a mirror vdev. Adding space to a pool can ONLY be done by tacking another vdev to the pool, so if your pool is RAID5, then you can only tack another RAID5 vdev (consisting of multiple disks) to the pool. You cannot simply "add a disk" to a RAID pool, and you can never remove a disk or vdev from a pool without completely destroying the pool.

    4. Re:Is that all? by kildurin · · Score: 2, Interesting

      And I would do RAID why? I only use JBODs with ZFS. I find it very strange to use RAID5 arrays with ZFS. We have basically replaced all our RAID5's with ZFS or left them entirely UFS. For most uses, a partition on the disk (usaully just one for the entire drive) is used as the vdev. Thus adding drives is the way to expand a pool.

    5. Re:Is that all? by ms139us · · Score: 2, Interesting

      Sorry, I was unclear. If you have a RAIDZ pool (consisting of JBODs), you cannot simply add a drive to the pool. Instead, you must add another RAIDZ set to the pool.

      Suppose you start with a pool that is a RAIDZ set with 5x250GB SATA drives (4 data + 1 parity), giving 1TB of data

      3 years later, your needs have grown to 3TB and you need better performance and double parity. The right drives for you now are 1TB SATA4 drives. You really want to get rid of your old and slow SATA array, but your system is 24x7, so you cannot take it offline to do a backup.

      Your only option is to add an array of 3x1TB (2 data+ 1 parity) drives as a RAIDZ set to the pool and keep the old array around until you go out of business or take the system offline for a backup and you cannot move to double parity.

      If your original array gets so old that you can no longer find drives for it, your sunk.

  15. Is it too much to ask that the summary writer... by objekt · · Score: 1

    ...mentions that ZFS is from Sun?

    --
    -- Boycott Shell
  16. Switch all filesystems to ZFS... by athloi · · Score: 5, Interesting

    Once we're sure it's stable, because it looks like a massive improvement over the 1970s-style file systems we're using now. ZFS is now part of FreeBSD, Solaris will have ZFS "soon" and many Linux distros are also considering it. Good. Let's get to a common standard that's excellent and forget the tedium of these past, less effective file systems.

    1. Re:Switch all filesystems to ZFS... by target562 · · Score: 3, Informative

      Solaris having ZFS "soon"? Looks like an old link, as it's been part of Solaris 10 since last summer... My servers running it in production would be sad to hear if it wasn't...

    2. Re:Switch all filesystems to ZFS... by Anonymous Coward · · Score: 1, Interesting

      Hmmm... I've been running a linux laptop with SGI's XFS as my root filesystem for the last 5 years, that's a 1990's-era filesystem. Never had a problem with it, including surviving many kernel panics and random emergency power-offs. Seems to be a great filesystem for real engineering-type workloads (lots of files, big files, etc).

      From what I've heard, ZFS is being promoted over a much better (i.e. backwards-compatible) in-house filesystem by a bunch of ex-Sun zealots who now work at Apple...

    3. Re:Switch all filesystems to ZFS... by dave420 · · Score: 1

      They've got to iron out some of the performance issues first, but yeah - it would be cool.

    4. Re:Switch all filesystems to ZFS... by Mad+Merlin · · Score: 1

      ...and many Linux distros are also considering [ZFS].

      Consider is the most they'll do. ZFS's license is incompatible with the GPL, and thus will not be integrated into Linux. Some would say that Sun deliberately choose a GPL incompatible license for ZFS, but that's a topic for another flamewar.

    5. Re:Switch all filesystems to ZFS... by Tsunayoshi · · Score: 2, Informative

      You linked to an article from 2005...ZFS has been in Solaris 10 since update 2 (06/06). New features, enhancements, and optimizations appeared in update 3 (11/06). It just will not be available as a booting FS until sometime in 2007.

      The OpenSolaris project is currently working on getting bootable ZFS support (available in the current release as experimental).

      --
      "Get a bicycle. You will not regret it, if you live." - Mark Twain, "Taming the Bicycle"
    6. Re:Switch all filesystems to ZFS... by AnyoneEB · · Score: 1

      I have seen this mentioned a few times. Although not optimal, would it be possible for a group to simply read the docs, reimplement ZFS, and release the result as GPL-compatible code? The real question here is one of how open the docs are.

      --
      Centralization breaks the internet.
  17. Let's hope it's the truth by Otterley · · Score: 3, Insightful

    If ZFS is the default file system, it will mean that Time Machine (i.e. the snapshot feature) of 10.5 will be able to take snapshots without requiring a secondary file system to keep the copied (recoverable) blocks, as it does now with HFS+. To me, the secondary filesystem requirement makes Time Machine essentially useless on a laptop.

    1. Re:Let's hope it's the truth by larkost · · Score: 3, Informative

      TimeMachine is a backup tool, not really a live versioning tool. That makes having a second volume a requirement. If you don't understand that, then you don't understand what backups are for.

      I already know how TimeMachine is going to work (it was part of the filesystem presentation at last years WWDC... so I know it, but can't reveal it), and unless they have completely redone that entire system (which was quite elegant), then ZFS will not bring a single thing to it. I do know how ZFS could make that all really elegant, but Apple already has it covered on HFS+.

    2. Re:Let's hope it's the truth by Otterley · · Score: 1, Troll

      On the contrary, Time Machine is not a backup mechanism. Because it uses copy on write (i.e. only stores changed blocks), you will not be able to restore your entire disk from the secondary volume (unless, of course, it requires the target volume be larger than the source volume and does a quiescent backup of the entire source volume before it kicks in).

      For this reason, the secondary volume requirement does not seem particularly useful.

    3. Re:Let's hope it's the truth by beezly · · Score: 2, Insightful

      TimeMachine is a backup tool, not really a live versioning tool. That makes having a second volume a requirement. If you don't understand that, then you don't understand what backups are for.

      I can think a few good cases for "backing up" to the same physical device.

      Here are the reasons I back up data my employer's data:

      • Hardware failure. Disks die, tapes fail, etc. RAID helps guard against this, but it doesn't help you guard against disks that fail silently (ie, they corrupt data rather than return an error). Backups to the same device are generally useless in this case.
      • Software failure (bugs). Your OS writes garbage to the filesystem, or your application writes garbage. Backups to the same device can sometimes protect against data-loss in this case.
      • User error. User deletes files. Backups to the same device are good in this case.
      • Operator error. I delete files. Of course, this never happens! If it ever did, backups to the same device would probably mean I could restore the data quickly.
    4. Re:Let's hope it's the truth by Wesley+Felter · · Score: 1

      RTFM.

      "Right from the start, Time Machine in Mac OS X Leopard makes a complete backup of all the files on your system. That includes your system files, applications, accounts, preferences, music, photos, movies, documents -- everything you keep on your Mac."

    5. Re:Let's hope it's the truth by Otterley · · Score: 1

      I stand corrected. However, if this is the way it works, the implementation is seriously lame. Real snapshot capability has been a viable filesystem feature for over 10 years, and apparently Apple still can't get it right.

    6. Re:Let's hope it's the truth by Wesley+Felter · · Score: 1

      Snapshots don't protect against disk failure and they are non-trivial to implement. Apple chose a simpler implementation (copying files) that combines most of the benefit of snapshots and backups.

    7. Re:Let's hope it's the truth by Otterley · · Score: 1

      So it sounds like the scheme is like master-slave replication a la MySQL.

      What happens if the target filesystem is disconnected, and then reconnected? Do you lose all the intermediate changes that occurred in the meantime?

      Moreover, what's the recovery mechanism using the secondary filesystem once the OS has been reinstalled on the primary disk? This is clearly not a mirrored filesystem that can simply be booted from.

    8. Re:Let's hope it's the truth by Anonymous Coward · · Score: 0

      This is clearly not a mirrored filesystem that can simply be booted from. No, you can't boot from TimeMachine backups, but there are numerous other ways to use them.

      They appear in Finder as a bunch of directories with complete copy of backed up disk in each one (directory hard links are used to reduce actual disk space used) So you can just drag and drop individual files (eg. The version of that doc I was working on last night) An ACL is used to make the backups readonly to prevent accidental modification of the backup.

      DiskUtility can do a restore from a TimeMachine snapshot

      MigrationAssistant can transfer users/data from a TimeMachine snapshot
    9. Re:Let's hope it's the truth by Anonymous Coward · · Score: 0

      The Leopard install disk can restore from a Time Machine backup. That's how you recover from a hosed OS drive.

  18. Leopard secret features by Anonymous Coward · · Score: 0

    I think this is true. ZFS is maybe the biggest of Leopard secret features that will be revealed next week at the WWDC. I believe this is the real reason Leopard was delayed.
    If it's really the default file system on OS X 10.5 it really is big news.
    We will see.

  19. OT Grammer/Spelling Nazi Alert by multisync · · Score: 4, Insightful

    because ZFS has back-up and snapshots build right in to the filesystem


    I think Slashdot would benefit from adopting some of K5's approach to story submissions. The Firehose is a great start, but instead of simply saying yes or no, users should be able to give feedback to the submitter. The summary for this article is a great example. The submitter typed "build" instead of "built," resulting in an annoying distraction in an otherwise concise description of the story.

    Newspapers have Copy Editors (at least they used to; most seem not-too-bothered by spelling these days). It would be nice if interested Firehose users were given the opportunity to help make sure the summary was fit for publication before it hits the front page.

    I guess this should have been a journal entry, but it seemed like an opportune time to bring this up.
    --
    I don't care why you're posting AC
    1. Re:OT Grammer/Spelling Nazi Alert by doombringerltx · · Score: 1

      I agree. The other day on the firehose I saw a summary refer to "Huxley's 1984" and my soul cried a little bit.

    2. Re:OT Grammer/Spelling Nazi Alert by g0at · · Score: 1

      You misspelled "grammar".

      But I do fully agree with the essence of your argument. :)

      -b

    3. Re:OT Grammer/Spelling Nazi Alert by QRDeNameland · · Score: 1

      The submitter typed "build" instead of "built," resulting in an annoying distraction in an otherwise concise description of the story.

      For whatever it's worth, I submitted a story a few weeks back about author Mark Helprin's argument for perpetual copyright, and even though I am certain that I spelled his name correctly in my submission, the summary had it misspelled "Halprin". And yes, the summary notes that several readers submitted the same story, but most of the summary used my exact wording, yet they still spelled the name wrong.

      My point being...sometimes the errors come from the *editors* and not from the submitter.

      --
      Momentarily, the need for the construction of new light will no longer exist.
    4. Re:OT Grammer/Spelling Nazi Alert by andrewd18 · · Score: 1

      You're supposed to put periods inside quotation marks, not outside.

    5. Re:OT Grammer/Spelling Nazi Alert by MightyYar · · Score: 1

      Only if it is a quote, IIRC.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    6. Re:OT Grammer/Spelling Nazi Alert by kv9 · · Score: 1

      You're supposed to put periods inside quotation marks, not outside.

      bullshit.

  20. Video Demo of ZFS by kildurin · · Score: 3, Interesting

    I know its not on Mac but this shows how easy and powerful ZFS is. I have heard directly from Sun that by Solaris 10 will soon have bootable ZFS either in update 4 or update 5. Remember that the big problem with Sun hardware is that they need firmware support for bootability and that it may be much easier on OS X to make ZFS bootable. http://video.google.com/videoplay?docid=8100808442 979626078

    1. Re:Video Demo of ZFS by chrome · · Score: 1

      that video makes baby jesus cry.

  21. This can, potentially, make upgrades a pain.. by N3WBI3 · · Score: 3, Insightful

    This will make going from earlier versions of OSX to the new one more of a pain because the whole disk will have to be reformatted.

    --
    1. Re:This can, potentially, make upgrades a pain.. by kildurin · · Score: 0

      No, ZFS does not require reformatting. It really is about a two second command to create even on the largest Jbod's I have (over a terabyte in 250G drives). I also do not believe that Apple will force it on you.

    2. Re:This can, potentially, make upgrades a pain.. by N3WBI3 · · Score: 1

      Thanks, I was under the impression going from HFS to ZFS was not so trivial. I thought it was more like going from fat32 to ntfs and not like going from ext2 to ext3.

      --
    3. Re:This can, potentially, make upgrades a pain.. by Anonymous Coward · · Score: 0

      Apple doesn't force things on people, they move on. I don't remember backward compatibility ever being a concern for Apple (hardware or software).
      What about the meta-data and permissions from their current file system? Will it transfer cleanly to ZFS? I bet not.

    4. Re:This can, potentially, make upgrades a pain.. by maubp · · Score: 1

      A few seconds? Surely all that can be doing is setting up an empty ZFS file system on a new drive (overwritting any existing partition). I guess that's still much faster than other file systems.

      However, wasn't the OP asking about converting an existing HFS+ partition into a ZFS partition while preserving the data - that would surely take a significant amount of time.

      I expect it would be optional for upgraders, who might just stay on HFS+

    5. Re:This can, potentially, make upgrades a pain.. by dabraun · · Score: 1

      Thanks, I was under the impression going from HFS to ZFS was not so trivial. I thought it was more like going from fat32 to ntfs and not like going from ext2 to ext3.


      Windows has long included tools to convert fat/fat32 volumes to ntfs in-place (both during upgrade and after the fact). I'm sure Apple can pull this off if they want to.
    6. Re:This can, potentially, make upgrades a pain.. by larkost · · Score: 2, Interesting

      No, you are wrong. Going from HFS+ to ZFS would require some tricky maneuvers to get the data moved over. On the Intel Macs (actually... on Mac's formated with EFI's disk format, but these are only Intel Macs) can volumes be dynamically resized, which would be needed in this case. Even then there would be some real gymnastics involved on a disk over 50% full.

      You are right that ZFS can handle volume size changes live (and HFS+ can sort-of do it), but this does not mean it is a slam-dunk. I would not want to be a product manager in charge of providing the transition code.

    7. Re:This can, potentially, make upgrades a pain.. by N3WBI3 · · Score: 2, Insightful

      And the windows tool has *always* been crap which results in less stable and predictable systems.

      --
    8. Re:This can, potentially, make upgrades a pain.. by nine-times · · Score: 1

      Yeah, but it shouldn't be that much worse, really. You're only getting this if you buy a new Mac or seek out a Leopard upgrade. In the second case, you should have some idea of what you're doing anyway, and you should always backup your data before doing a major OS update. (you should be doing regular backups anyway)

      IMO, if you lose much by reformatting your hard drive, you're using your computer wrong. You should always have some kind of backup, keep disks and serial numbers for all your software, etc. Any decent OS will keep all your configuration files in sensible places, too, and therefore easy to backup. In the case of OSX, dragging your home folder onto an external harddrive is enough for most people.

  22. You misspelt "grammar" by Anonymous Coward · · Score: 0

    Back, nazi, back!

    1. Re:You misspelt "grammar" by Anonymous Coward · · Score: 0

      Maybe I need Firehosers to vet my comments while they're at it ;)

  23. Two things... by ErikZ · · Score: 1


    How will ZFS work with Bootcamp?

    When is Leopard coming out? I know it's been delayed, maybe they're testing it with ZFS?

    --
    Democrats or Republicans. They are both taking us to the same place and they are not afraid of us anymore.
  24. case-insensitive: performance, i18n, safety by r00t · · Score: 1, Insightful

    Performance:

    Suppose I want to access a file.

    First, the filesystem looks it up. This operation takes time proportional to the log of the directory size. Maybe you do better with hashes.

    On a case-sensitive (POSIX-compliant) filesystem, you're done. You have the file, or you can return an error code.

    On a case-insensitive filesystem, your done if you're lucky. If not lucky, you need to do a linear scan of the whole damn directory. Many places have a directory with some insane amount of files. Intentionally or not, it's common to go into the tens of thousands. A few places (running XFS mainly, sometimes Reiserfs) get into the millions.

    Because of the way directory listings are done (read then look up stats) you can generally square the above numbers. Ouch.

    I18N:

    Then there is the issue of internationalization. For example, consider "I" and "i". Some places have an uppercase with the dot, and other places have a lowercase without the dot. The rules for uppercasing and lowercasing differ from what most people are used to. Oh crap! This issue doesn't exist on a case-sensitive filesystem.

    Safety:

    App needs to make a file. App sees that file does not seem to exist. App writes file. Complex international case rules mean that no, the file DOES exist, and it gets clobbered.

    1. Re:case-insensitive: performance, i18n, safety by Fweeky · · Score: 1

      "On a case-insensitive filesystem, your done if you're lucky. If not lucky, you need to do a linear scan of the whole damn directory."

      Er, why? If you're case-insensitive, you use a casefolded index for lookups, surely?

    2. Re:case-insensitive: performance, i18n, safety by kithrup · · Score: 5, Informative

      Sorry, but your description of the lookup process isn't right. First: lookup depends on the way directory entries are store. On UFS, it's a non-sorted array; in order to do a lookup, you need to (worst case) scan the entire directory. On VxFS, they use a hash, so first you hash the input, and then do run through the entries that have a matching hash. On HFS+, the catalog is stored as a B-tree, so you do compares to get to the right node, and then look through the node until you either find it or reach the end of the node. Second: None of those is affected by case-insensitivity. You simply do a case-insensitive compare each time. This is the difference between HFS+ and HFSX on Mac OS X: in the former, the key-compare function is a Unicode case-insensitive comparator; on the latter, it's just a memcmp. Third: your comment about "i" is a glyphing issue, not a character issue. Apple has a pretty good technote up on their HFS+ impelementation, and it describes the way the case insensitivity works. I recommend reading it.

    3. Re:case-insensitive: performance, i18n, safety by Anonymous Coward · · Score: 1, Informative

      Here's a link to it: HFS Plus Volume Format

      I likewise highly recommend it.

    4. Re:case-insensitive: performance, i18n, safety by netfunk · · Score: 3, Insightful
      (I don't know what various filesystems actually do, this is just how I would assume it's done, at least on systems designed for case-insensitivity...ext2 or FFS probably would suffer from the issues you mention about scanning the whole directory.)

      On a case-insensitive filesystem, your done if you're lucky. If not lucky, you need to do a linear scan of the whole damn directory. And yet Windows and Mac OS have had case-insensitive filesystems for years and somehow they are usable, even with Unicode filenames.

      You can't restore the original case of a string afterwards, but you can always make it lowercase. This is called "case folding." You can fold two strings to a lowercase form, and then compare them for equality or whatnot. Works with Unicode, too.

      Then there is the issue of internationalization. For example, consider "I" and "i". Some places have an uppercase with the dot, and other places have a lowercase without the dot. The rules for uppercasing and lowercasing differ from what most people are used to. Oh crap! This issue doesn't exist on a case-sensitive filesystem. While folding Unicode chars is frequently presented as an unsolvable problem ("what do you do with the letter with the squiggly thing above it? Or converting that German capital 'B' thing to two lowercase 's' chars? There are MILLIONS OF THESE!") ... there are actually very few cases in the grand scheme of things. Most languages don't have upper and lower case, after all.

      Here's the whole list of characters that need to "folded" to a lowercase form, accounting for instances where it will cause the string to grow (like that German 'B' thing):

            http://www.unicode.org/Public/3.2-Update/CaseFoldi ng-3.2.0.txt

      (And you can hash those chars too, so folding a string doesn't involve hundreds of conditionals.)

      If you don't care about Unicode, case folding an English ASCII char is 2 lines of C code, and a few more if you want extended ASCII.

      Once you have a filename, you can store it in the filesystem as the specifically-entered characters, so you don't lose the original casing, but also store with it a hash of the case-folded version. Now whenever you need to look up a specific filename, you case-fold it, hash that folded string, and look it up that way against the hash you previous calculated when creating the file. Now it's as fast as the case-sensitive filesystem, minus the overhead of folding a small string.

      Because of the way directory listings are done (read then look up stats) you can generally square the above numbers. Ouch. The way directory listings are done doesn't change...readdir() is the same in all cases, and your lookup is still a hash. If you had to scan, the first run is slow anyhow due to disk bandwidth and seek speeds, but then a modern OS can cache the inodes to speed this up for the next run.

      App needs to make a file. App sees that file does not seem to exist. App writes file. Complex international case rules mean that no, the file DOES exist, and it gets clobbered. I would think that stat(filename) would not report the file doesn't exist if open() would then clobber it, at least not for case-sensitivity issues.

      If your app decides about a file's existence by using readdir() until it finds it, and doesn't properly case-fold, and didn't call open() with O_EXCL, then not only did you go the long way about it, you got what you deserved for clobbering the file.

      Actually, if you don't just open(O_CREAT | O_EXCL) to check for existence and create if missing in one step, then you'll have an atomicity problem anyhow. Use the services the OS provides, they are there for a reason.

      --ryan.

      --
      Don't say, "don't quote me," because if no one quotes you, you probably haven't said a thing worth saying.
    5. Re:case-insensitive: performance, i18n, safety by gnasher719 · · Score: 1

      I don't think you have ever programmed using HFS+. The problems that you imagine just don't exist.

      In a case-sensitive file system, or to be more precise, in a file system where any file names containing two different letters are considered different, you compare filenames by comparing individual letters. In HFS+, each letter is mapped to a code using some table lookup, and then the codes are compared to see if two filenames are different or not, and how they are ordered; some characters are even ignored. This makes the comparison code slightly more complicated, but not much. The number of filename comparisons is unchanged.

      Second, there is no issue with internationalization at all. In HFS+, the tables used in the table lookup described above are independent of any locale, and they are independent of any changes in the Unicode standard as well. HFS+ doesn't care about internationalisation at all. And since every bit of code that looks for the existence of a file uses the same algorithm (for example "stat" which one might use to determine whether a file exists, and "fopen" which also needs to determine whether a file exists), your last scenario cannot happen.

    6. Re:case-insensitive: performance, i18n, safety by r00t · · Score: 1

      NTFS filesystems actually have a case translation table stored on the disk. The one for the USA does not match the one for Turkey.

      Apple might not give a shit, in which case HFS+ behaves wrongly in Turkey.

  25. Harsh reply! by grandmofftarkin · · Score: 2, Funny

    There is no need to 'stick the knife' in! ;-)

  26. Re:Is it too much to ask that the summary writer.. by Anonymous Coward · · Score: 0

    I'd never heard of ZFS before, but the summary made it pretty obvious to me that it was from Sun (an announcement from the Sun CEO, and a prior request that Sun port it for them).

  27. Whither Macintosh HD?? by wandazulu · · Score: 1

    This is just total utter complete speculation on my part, but one of the features of ZFS I like is the pool concept where you just add disks to the total storage space and then carve it up any way you see fit. I tested this with Solaris 10 on VMware and it really is just that easy...add the disk to the machine, then register it with the pool. Done. No muss, no fuss. I was shocked at how fast and easy it was.

    Anyway, where I'm going with this is: Why should we still have disks on the desktop? If all the physical disks are just part of one big pool, you do away with needing to see the various disk icons and instead have something maybe similar to "This computer" with a different organizational structure; you see Applications as a group (a la the original Windows Program Manager) while your home directory stays the same in the nested-folder style.

    What I'm really wondering is if ZFS could be an integral part of a totally redesigned Finder interface that allows for Apple to get away from the traditional Disk/Directory paradigm. I know Unix in general really doesn't support a disk paradigm at all (which is why I've always preferred Unix over Windows...none of the C: or D: business) and it'd be great to have the same thing on the Mac with Apple's capability for superior interfaces.

    1. Re:Whither Macintosh HD?? by Hatta · · Score: 1

      This is just total utter complete speculation on my part, but one of the features of ZFS I like is the pool concept where you just add disks to the total storage space and then carve it up any way you see fit.

      How exactly does this work with respect to redundancy? If I throw 2 500gb drives and a 200GB drive into the pool is it going to be smart enough to figure out where all the redundancy needs to go so that if one of the 500gb drives dies I can repair everything? what if I have 1 500gb drive and a 200gb drive obviously I can't recover from a failure of the 500gb drive, is ZFS smart enough to figure this out and warn me?

      --
      Give me Classic Slashdot or give me death!
    2. Re:Whither Macintosh HD?? by wandazulu · · Score: 1

      It seems that they have some built-in redundancy:

      http://www.sun.com/emrkt/campaign_docs/expertexcha nge/knowledge/solaris_zfs_perf.html#28

      I have not tried this myself (though I suppose it'd be a cool experiment using VMware) but I'm guessing that ultimately you'd probably be SOL for the stuff on the disk, though as I understand it the pool would still be up and running, with some directories and files unavailable.

    3. Re:Whither Macintosh HD?? by gobbo · · Score: 1

      Why should we still have disks on the desktop? If all the physical disks are just part of one big pool, you do away with needing to see the various disk icons and instead have something maybe similar to "This computer" with a different organizational structure; you see Applications as a group (a la the original Windows Program Manager) while your home directory stays the same in the nested-folder style.

      I like to have discrete disks (or arrays) because I have different tasks for them. I apportion out their use based on performance and expected reliability. Perhaps my attitude will be moot shortly as drives and busses speed up and become more reliable, but for now it's sensible.

    4. Re:Whither Macintosh HD?? by atrus · · Score: 1

      For the reasons you stated, ZFS doesn't try automagically making things fault tolerant, leaving it up to the human to figure out.

      So when you add disks, you add them in mirrors, raidz (raid5ish), or raidz2 (raid6ish) vdevs. The trick is added volumes automatically become part of the underlying storage pool, and data gets striped over these volumes. You can also upgrade capacity by systematically removing drives from a mirror/raidz vdev, waiting for a recover, and repeat. Once the smallest drive in the mirror/raid vdev are replaced, the capacity instantly increases.

      It is possible to have a pool of mixed mirrors and raidz sets, but it is not recommended (differing levels of redundancy - as data is striped over all the sets the weakest point can fail everything).

  28. By saying "THE" filesystem... by csoto · · Score: 1

    ...The Schwartz meant "THE" as in "it's THE filesystem all the COOL KIDS will be using." He didn't necessarily mean the ONLY filesystem.

    Oh, and can you imagine a beowulf cluster of these things?

    --
    There exists no way of exchanging information without making judgments. --Bene Gesserit Axiom
    1. Re:By saying "THE" filesystem... by bvimo · · Score: 1

      Oh, and can you imagine a beowulf cluster of these things?
      I certainly can and it's getting me quite aroused.
      --
      In either case, here at Microsoft, we feel standards are important. And we have fun, too. Doug Mahugh, Microsoft
  29. Boot Camp does by SuperKendall · · Score: 1

    can volumes be dynamically resized, which would be needed in this case

    The Boot Camp setup does exactly that (shrink existing partitions on Intel macs).

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
    1. Re:Boot Camp does by freedumb2000 · · Score: 1

      Yes, shrink. And only that, it's really not that spiffy. It requires youto have completely free space at the end of your partition. All it does is rewrite the partition table to end at an earlier position. On any normaly used disk you will have to do full fragmentation first before doing any shrinking. So even doing something as simple as setting up boot camp requires a very specigic setup.

  30. Other benefits by SuperKendall · · Score: 1

    It's easy to add custom filesystems to ZFS, so you could have something like a filesystem that represented a remote photo sharing site as a directory (that's kind of a lame example, but I hope it serves to illustrate the ability).

    Also ZFS supports automatic low level compression, so disk reads/writes can be faster since you are reading compressed data off a disk (disk reads are a lot slower than decompression).

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
  31. lets look at the bigger picture here by hackstraw · · Score: 1


    I'm a new ZFS user, and so far I'm very happy with it.

    But the bigger picture here is that this is NOT a YRO.slashdot.org article. Its just a news article.

    No patent disputes. No lawsuits. Just technological progress and cooperation.

    Why can't this be more of the status quo?

    I've been reading stuff about predictions by scientists, and its still universal that hardware will increase at the current rate of about doubling in power ever 12-18 months.

    Software performance is about 1/2 of that at best. If more cooperation like this could happen, then there is no reason for software to lag behind hardware, but one large software company keeps hindering this progress...

    1. Re:lets look at the bigger picture here by sussane · · Score: 0

      where are the features of ZFS, if its so gooood.

      --
      Best Regards, Eliena Andrews
    2. Re:lets look at the bigger picture here by mink · · Score: 1

      So far it seems to be faster and makes better use of the drive array (SUN A1000) then Windows Server 2000 did with RAID5. The added bonus of raidz2 being able to survive 2 failed disks is nice. The setup (even though I had to compile stuff) was better/easier then I expected.

      Those are the features I am using.
      There are other like snapshots and whatnot that I am not using currently but I'm sure other can share that.

      root@linuxfc:/# zpool status
          pool: share
        state: ONLINE
        scrub: none requested
      config:
      NAME STATE READ WRITE CKSUM
      share ONLINE 0 0 0
          raidz2 ONLINE 0 0 0
              sda ONLINE 0 0 0
              sdb ONLINE 0 0 0
              sdc ONLINE 0 0 0
              sdd ONLINE 0 0 0
              sde ONLINE 0 0 0
              sdf ONLINE 0 0 0
              sdg ONLINE 0 0 0
              sdh ONLINE 0 0 0
              sdi ONLINE 0 0 0
              sdj ONLINE 0 0 0
              sdk ONLINE 0 0 0
              sdl ONLINE 0 0 0
      spares
              sdm AVAIL
      errors: No known data errors
      root@linuxfc:/# zpool list
      NAME SIZE USED AVAIL CAP HEALTH ALTROOT
      share 394G 140G 254G 35% ONLINE -

      There should be an sdn device but one of my drives failed and I cant find my 1 extra disk (more are on order) that was not in the array. When I find it there will be a second spare device.

      I managed this all under linux. I am thinking I will migrate the system to Opensolaris once I ensure my LSI HBA has support.

      --
      Well I've wrestled with reality for thirty five years doctor, and I'm happy to say I finally won out over it.
  32. ZFS by C_Kode · · Score: 1

    I think ZFS on Mac is a great idea, and I'm not even an Apple Fanboy. What I want is ZFS ported to Linux, (not via FUSE) but that would require Sun to change or dual license ZFS to allow it. (Currently under the CDDL license I beleive)

    1. Re:ZFS by _Knots · · Score: 1

      You are free to reimplement ZFS for Linux, not in FUSE. You might even start from the GPL'd code in GRUB at http://src.opensolaris.org/source/xref/onnv/onnv-g ate/usr/src/grub/grub-0.95/stage2/ .

      --
      Anarchy$ dd if=/dev/random of=~/.signature bs=120 count=1
  33. The best of Unix? by gilesjuk · · Score: 4, Interesting

    Seems like Apple take some of the best ideas from the Unix world. Really shows the potential of Unix systems if the people who wrote them thought a little more about usability.

  34. will ZFS serve as a replacement for Xsan? by blakieto · · Score: 1

    I'm interested in hearing if ZFS would effectively replace Xsan. Anyone know?

    1. Re:will ZFS serve as a replacement for Xsan? by Anonymous Coward · · Score: 0

      No. ZFS is not [currently] a "cluster" file system.

    2. Re:will ZFS serve as a replacement for Xsan? by t35t0r · · Score: 1

      Yes, I'm curious about this as well. We just purchased lots of raid boxes and SAN switches to setup several TB in XSAN. Will ZFS be used in the XSAN as the new filesystem or will XSAN slowly be deprecated in favor of some sort of ZFS setup?

    3. Re:will ZFS serve as a replacement for Xsan? by ChrisA90278 · · Score: 1

      ZFS is software. Software no matter how good can't replace a fiber channel cable or switch. I think ZFS will be the overall framework into wich Apples storage fits. So you would "add a RAID to a Pool" in the new terminology.

    4. Re:will ZFS serve as a replacement for Xsan? by t35t0r · · Score: 1

      We're not asking it to replace the hardware but become the FS for the SAN. This means all the snapshotting and nice features are part of the SAN.

  35. ZFS for FUSE by piojo · · Score: 3, Informative

    I am using ZFS-FUSE right now. On my gentoo system, many partitions are zfs, including /home, /var/tmp, /usr/share, /usr/portage, and /opt

    Because I have suffered some random corruptions in the past, even with ext3 ("This mp3 didn't used to have a skip there!"), I wanted the checksumming so that I can tell when I need to restore something from a backup.

    As a filesystem, it works completely, including creation of new filesystems, compression, checksums, etc. However, I've noticed a decrease in my system's general performance since installing zfs (probably due to it holding my home directory). Memory usage and mysterious CPU usage (I don't think it's checksumming) are the current disadvantages, but the author says it's still completely unoptimized.

    Should you try zfs-fuse? Definitely. But right now don't expect a performance gain.

    --
    A cat can't teach a dog to bark.
    1. Re:ZFS for FUSE by morcego · · Score: 2, Informative

      I've noticed a decrease in my system's general performance since installing zfs


      A user space filesystem is not something I would expect performance from.
      --
      morcego
  36. Um, you don't actually know that... by Anonymous Coward · · Score: 0

    You have know way to know if bits in your movie files are being corrupted with FAT32 unless you manually keep around checksums. ZFS does that for you. It will detect silent corruption of your data.

  37. Apple breaks in stages, default in 10.6 by alexhmit01 · · Score: 4, Insightful

    Apple is "well known" for massive backwards compatibility updates... except they aren't... They always handle transitions over a couple of versions, intelligently bringing people along. They swapped processor architectures twice and each time brought people along with emulators, in the Intel case it wasn't faster than the fasted G5 machines, but those of us upgrading 3+ year old machines (Powerbook G4 1Ghz -> Macbook Pro in my case) found our PPC apps running faster and Intel code flying.

    We all expected the Intel migration to happen with 10.5, they shocked us when they did it off the 10.4 base.

    While they did abandon Mac OS to move to OS X, they provided a migration strategy (Carbon) and a compatibility layer (Classic). Classic support shipped with 10.0/10.1, 10.2, and was supported in 10.3 if you already had it, as well as 10.4 I think, but they kept classic for around 5 years, which gave everyone time to migrate to Carbon. Its unfortunate that there is no long-term Classic via Rosetta just from a classic application point of view, but they didn't leave anyone in the lurch.

    I expect 10.5 to introduce this OS, which will be useful for new installs, or for external drive arrays, especially for the Video market, but I wouldn't expect it to be the default. OS X has supported a Unix filed system, but defaulted to HFS+, because HFS+ was compatible with Mac OS, so you could dual-boot OS 9 and OS X for a good 2 years on new hardware to maintain compatibility. If they hadn't done that, they would have lost the Pro-Audio and Pro-Video markets that took a few years to get native OS X applications.

    Getting it in the wild and for professionals would help that market, while not breaking ANYONE's compatibility. Sometime in 10.5's lifetime they may ship new computers with it, or they may wait for 10.6 in two years. But giving everyone two years is plenty of time to get utilities and applications compatible with the new file system.

    The flashy consumer features are touted for the OS, but the underlying architecture has always followed a 2-cycle release. If you've used OS X Server for 10.2/10.3/10.4, you'd notice that they introduced stuff in one version with limited exposed functionality (with the rest via the Unix layer), enhanced the functionality in the next rev, and polished thereafter.

    The Apple Mail Server -> Cyrus migration was someone poorly handled, but mostly because AMS was garbage. But the 10.4 Mail tools are night and day beyond the 10.3 ones.

    They are actually far more careful than people give them credit for.

    The different is, they don't keep backward compatibility as a long-term goal, they do a two-stage migration, giving people 2-4 years to transition.

    1. Re:Apple breaks in stages, default in 10.6 by Anonymous Coward · · Score: 0

      Can't the old time Mac savers on FAT systems such as (really working) DS_Store and various files which emulates resource forks come as saver for compatibility?

      Remember they were problem on FAT since it is massive fragment prone filesystem which has only single flag, not like ultra modern ZFS.

      There are 2 things I am afraid to see on OS X: /usr/local/bin/itunes/7.2.2/itunes and the ugliest hack ever, ".ico" files all over the place while /Applications only have .lnk files.

      Yes, I am afraid of going back years and years just to make couple Unix purists feel "elite" or on NeXT workstation. Imagine the horrible disk with tiny files all over the place.

      Remember Apple sells a new generation NeXT with Mach and FreeBSD hybrid running at background. It sounds like geekest fantasy ever but they manage to sell it to 60 year old ladies just because it is "easier" and "simple".

      They also manage to sell it to nerds who will happily run their X11 and Aqua as bonus. That is some excellent but fragile mix.

      (sorry for AC, really public terminal here)

  38. Sun is the new Bell Labs. by teknopurge · · Score: 2, Insightful


    Discuss.

    1. Re:Sun is the new Bell Labs. by chrome · · Score: 1

      I agree.

    2. Re:Sun is the new Bell Labs. by Khakionion · · Score: 1

      "Discuss."

      There's this huge stain in the basement of my school's building. What is that?

      --
      OMG! Wau!
  39. What I Want To Know Is... by SwashbucklingCowboy · · Score: 1

    Did Apple use ZFS under CDDL or did they negotiate a custom license with Sun, i.e. is Apple's ZFS implementation open source or not?

  40. Triple Booting by FunkyELF · · Score: 1

    So now if I were to get a Mac and triple boot Windows OSX and Linux...

    Linux will only be able to read the Windows NTFS partition.
    Linux won't be able to do anything with the OSX partition (unless I use FUSE).
    Windows won't have access to anything other than itself.
    And I'm not sure about OSX, I've never used it.

    Is that correct?

    1. Re:Triple Booting by Tatsh · · Score: 1

      So now if I were to get a Mac and triple boot Windows OSX and Linux...

      Linux will only be able to read the Windows NTFS partition.
      Linux won't be able to do anything with the OSX partition (unless I use FUSE).
      Windows won't have access to anything other than itself.
      And I'm not sure about OSX, I've never used it.

      Is that correct?


      First of all, did you forget the other formats Linux can read?
      Why NOT use FUSE?
      Windows could (although I don't recommend) use MacDrive to read the Mac partition (maybe ZFS in the future), and there is IFS for ext (and others I believe) partitions. You can also use Explore2fs to read the Linux partition (not write to ext3). I don't recommend IFS very much either. It does not respect permissions.
      OS X has drivers to read NTFS, not sure about ext or Reiser and such.

      My last question is: have you ever used Linux?

    2. Re:Triple Booting by SteelWheel · · Score: 1

      Are you referring to right now, or if/when OS X switches to ZFS? With the state of play as it is today: Windows doesn't have access to anything but itself natively, but you can use MacDrive for Windows from Mediafour to give Windows the ability to read/write from HFS volumes: http://www.mediafour.com/products/macdrive/

  41. Your Geek License is officially revoked! by Anonymous Coward · · Score: 0

    go back to Digg

  42. As much as I want ZFS on my laptop.... by Above · · Score: 1


    I'm not sure a boot device / standard first partition for a laptop is where ZFS shines, or where Apple would put a lot of focus. However, look at the server space. Lots of people like XSan. Products like Final Cut Server demand huge piles of storage, often XSan, and could use a lot of the properties of ZFS file systems. They could also be exported directly as file systems and mounted on desktops and laptops.

    "Default" in this case may mean the default for your XServe RAID, or XSan, but not for your laptop.

  43. Submitter agrees by Fjan11 · · Score: 1

    I agree. I am the submitter of this story btw. Something weird happened. Only the first half of the description is mine, the other half was taken form another firehose article. My closing line was "Jobs probably doesn't like someone stealing his thunder for the keynote" and it doesn't appear in the current summary. But oddly enough some people in the discussion are referring to that line, so this line must have been taken out after the story was already up...
    I did make the d/t error though. I am a dyslexic and a non-English speaker, so you could say I'm double handicapped. ;-)

    --
    This sig is just as redundant as the rest of this posting
    1. Re:Submitter agrees by multisync · · Score: 1

      That's interesting. This comment mentions spelling errors being introduced in to the summary of an article he submitted as well. I guess it makes sense. Lot's of people submit the same article, and it must be tempting for the editors to cut a bit out of this one and graft it on to that one. Especially if it improves the quality of the summary. Of course, a user might object to his name being associated with the comments of another submitter, but I guess that's the risk you run submitting articles.

      BTW, you're doing pretty well for a dyslexic non-english speaker. I couldn't even manage to spell "grammar" correctly in my subject line, and I've been speaking and writing english exclusively my whole life!

      --
      I don't care why you're posting AC
  44. ZFS still has bugs by mlheur · · Score: 5, Informative

    For something that's only a year or so old (production wise), I don't trust it worth shit.
    We run Netbackup Enterprise on Solaris 10 - during our last round of upgrades we installed ZFS on our disk staging storage units. It replaced VxFS. The way disk staging storage units (DSSUs) work in Netbackup, the disk is always near 100% full form a unix perspective. Basically, any time more disk is needed, the oldest image that has been copied to tape is expired from disk, thus freeing up more room. However, ZFS's most prominent bug from our perspective is that during periods of high activity, if all blocks become allocated, it becomes impossible to unlink(2) a file. This causes the application to no longer be able to make space for new backup images.

    Going down the shell, try to rm a file and it comes back: rm failed, disk is full.

    Well, if the disk is full, and you can't rm because the disk is full, how do you free up space?

    Sun's response, truncate an unnecessary file using 'cat /dev/null > /path/to/file', then, once you have some blocks free, rm works (so does unlink).

    Ok - so how do you tell a compiled application to truncate an unnecessary file before unlinking it? You can't! How can you determine what an unnecessary file is? If you delete the image before expiring it from the catalog you get errors when you try to expire, so you end up with catalog corruption.

    All in all, this is a problem that should never have been introduced, let alone still exist after months of sending trace outputs and reproducing it in multiple environments. ZFS isn't ready for the real world.

    1. Re:ZFS still has bugs by Guy+Harris · · Score: 2, Informative

      if all blocks become allocated, it becomes impossible to unlink(2) a file.

      The price of copy-on-write. Other systems with a copy-on-write file system can exhibit this behavior.

    2. Re:ZFS still has bugs by Anonymous Coward · · Score: 0

      Are you, perchance, using ZFS's snapshot mechanism? Trying to 'rm' a file on a disk that is packed full of snapshots will error out the exact same way. I'm guessing you're not, but the coincidence was cute.

    3. Re:ZFS still has bugs by ralph.corderoy · · Score: 1

      If Sun said "cat /dev/null > /path/to/file" they've really gone downhill since the 1980s when I last used Sun OS. Back then, they'd have known it's just ">/path/to/file" from the shell.

  45. ZFS not ready for prime time by SirNAOF · · Score: 4, Informative

    ZFS is not ready for prime time - at least not on Solaris.

    I setup ZFS on some SAN storage in a new system. The internal boot disks were mirrored UFS. When one of the HBAs fried, the SAN storage disappeared - and the system panic'd.

    Every reboot thereafter stopped in a panic. The ZFS subsystem panic'd the system at every boot when it couldn't find all its volumes. After calling Sun support, I found out that they need to do a massive code redesign to catch that issue, and it wouldn't be out for at least 6 months.

    I'm sure ZFS will be great - once they clean up these type of showstopper bugs.

    --
    Jeremy Baumgartner
  46. bad decision by nanosquid · · Score: 0, Redundant

    I think this is a bad decision for Apple. ZFS cannot be incorporated into the Linux kernel due to Sun's choice of license, and Solaris isn't widely used compared to Linux.

    Regardless of whether ZFS is actually "better" than other file systems (I don't think it is), choosing ZFS makes Macintosh even less compatible with the mainstream FOSS world than it already is.

    1. Re:bad decision by Anonymous Coward · · Score: 0

      Too bad. Apple will just leap forward with superior technology while people like you whine about how you can't use it.

    2. Re:bad decision by Anonymous Coward · · Score: 0

      Too bad. Apple will just leap forward with superior technology while people like you whine about how you can't use it.

      Well, there's a first time for everything, I suppose, but I wouldn't hold my breath. So far, all Apple has ever done is copy other people.

  47. Time Machine will not require ZFS by abhi_beckert · · Score: 3, Informative

    Time Machine is already fully functional (apart from a few gui glitches) in the current leopard developer builds, but ZFS isn't even available in Disk Utility (yet?). This doesn't mean ZFS won't be added at the last minute, but it certainly isn't required for Time Machine.

  48. Except that ... by toby · · Score: 1

    ... booting OS X on a Mac and booting Solaris 10 on a *** are completely different questions.

    I don't see why Apple couldn't complete this functionality on that schedule: They know their operating system and integrated hardware intimately, are not burdened by Sun's qualification requirements, and have had considerable time (including the post-iPhone extension) to work on it.

    5:1 they ship with ZFS boot.

    --
    you had me at #!
  49. ...why not? by SanityInAnarchy · · Score: 1

    Why would you possibly prefer HFS+ to... well... anything else that OSX might support? Except maybe FAT32...

    --
    Don't thank God, thank a doctor!
  50. short memories by toby · · Score: 2, Interesting

    they shocked us when they did it off the 10.4 base.

    It did not shock those of us who know that NEXTSTEP was transparently portable to at least four architectures.

    --
    you had me at #!
  51. ZFS is reliable AND has recovery tools. by SanityInAnarchy · · Score: 1

    After enough years of use, don't you think someone would've noticed if ZFS didn't come with the equivalent of fsck?

    I'll bet money that ZFS will be both more reliable, and easier to recover than HFS+, except that it may take a little while before someone has a recovery GUI for you. Probably not, though -- it'll probably be in Disk Utility in the first release.

    --
    Don't thank God, thank a doctor!
  52. News Flash Apple Buys Sun.... by Glasswire · · Score: 0, Offtopic

    ..."Steve says 'It seems like a good idea' "

  53. So what will the preferences panel ... by constantnormal · · Score: 1

    ... be like, to allow all the various ZFS configuration options to be managed, and yet be simple and (relatively) idiot-proof?

    Seems to me that the design and implementation of THAT would take longer than to insert ZFS as the primary filesystem.

  54. Setting yourself up by TopSpin · · Score: 1

    I wonder whether this is really more than a research project within Apple. There is evidence that Apple has been investigating ZFS, but you have to take any third party claims about the design of future OS X releases with a fist sized salt crystal. Don't be surprised if Apple waits a few days and issues a 'clarification' on the matter. Jonathan Schwartz has set himself up for an embarrassment and, frankly, he deserves a slap for being so naive. They'll deny it even if incorporating ZFS is their intent. The economy sized egos at Apple will not tolerate being both up-staged and obligated by others. Whatever amount less of a splash Apple gets to make on its own as a result of Jonathan's public demonstrations will be accounted for in full, plus interest.

    --
    Lurking at the bottom of the gravity well, getting old
  55. What is new with ZFS? by Anonymous Coward · · Score: 0

    I really dont understand the hype about ZFS. Isnt it just the basic LVM + filesystem that weve had in UNIX (well, not Solaris) for years? Say for example in True64 (advfs/lvm) or AIX (jfs/lvm). I can stripe and mirror at a PP level any way I want. Snapshots and copies. What does ZFS provide that LVM + JFS doesnt? Can someone in the know summarize for us lazy? :)

  56. consider in-memory data structures by r00t · · Score: 1

    You'll probably be hashing filenames. I guess you can do that all lowercased, which may be OK for Apple.

    That doesn't work for Windows, which supports both Win32 and POSIX. You need something two-level, first case-insensitive then case-sensitive. (there are worse alternatives) Ugh.

    All of this is with per-filesystem case conversion, because filesystems can come from different locales.

  57. "glyphing" issue? I think not. by r00t · · Score: 1

    In one locale, the lowercase of "I" is the same as the lowercase of "I" with a dot. Both have dots. In another locale, they are distinct.

    In one locale, the uppercase of "i" is the same as the uppercase of "i" without a dot. Both are a plain "I". In another locale, they are distinct.

    Thus:

    In one locale, two filenames are distinct. In another locale, they are not distinct. Oh crap. Now try sharing removable media or a network filesystem between the two locales.

    1. Re:"glyphing" issue? I think not. by kithrup · · Score: 1

      Yes, you are confusing glyphing with characters. Take a look at the Unicode character set some time.

    2. Re:"glyphing" issue? I think not. by r00t · · Score: 1

      Perhaps you'd better define "glyphing". I spend a damn lot of time at the unicode web site, and I've never seen a verb form of the word glyph.

      I assume you mean the graphical representation of a Unicode code point. In that case, NO.

      Unicode has an "i" without dot. It has a distinct code point from the regular "i".
      Unicode has an "I" with a dot. It has a distinct code point from the regular "I".

    3. Re:"glyphing" issue? I think not. by Lars+T. · · Score: 1

      How the hell does your case-sensitive file system work when "i" and "i" are infact different chars, and thus "tit" and "tit" are two different files?

      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

    4. Re:"glyphing" issue? I think not. by r00t · · Score: 1

      Huh?

      There are 4 distinct code points here. Two are uppercase, and two are lowercase. Two have dots on top, and two do not.

      On a case-sensitive filesystem, all 4 are distinct. (they happen to look different too) This is very simple, easy, reliable, etc.

      On a case-insensitive filesystem, some of these code points get confused. Worse yet, it varies by locale.

      The obvious thing would be to preserve the presense/absense of the dot when changing case. We don't do this, even in English. In English, we match up the dotted lowercase with the undotted uppercase. That leaves the other things with one-way case transitions to cause trouble.

      So, are the dotted and undotted lowercase letters the same? No, but in English they both are the same as the uppercase undotted one! So this isn't transitive. Dotted lowercase is undotted uppercase, and undotted uppercase is undotted lowercase, but dotted lowercase is not undotted lowercase!

    5. Re:"glyphing" issue? I think not. by Lars+T. · · Score: 1

      Translation: "Only my made up problems are real, your's are not." Thanks for proving my point.

      --

      Lars T.

      To the guy who modded me down from perfect to terrible Karma - Apple haters still suck

  58. Upgrade? Just use MigrationAssistant by Anonymous Coward · · Score: 0

    Leopard installer asks right at the start if you want to create a backup of your disk.

    So just plug in an external Firewire/USB disk, do the backup, install Leopard with a reformat of disk to ZFS. Then when the Installer asks if you want to copy data from another computer/disk, just point at the backup. You end up with a clean install on ZFS with all your users, apps, data copied over...

    Much simpler and less error prone than trying to come up with a way of converting HFS+ volume to ZFS

  59. If Apple wanted a better file system... by argent · · Score: 1

    If Apple wanted a better file system they would ALREADY be using UFS.

    I'm really sick of HFS+. Three times now I've had it eat its brains so bad that FSCK wouldnt fix it. I have NEVER, not in 20 years that I've been working as a network administrator, had anything even vaguely resembling a modern UNIX file system get so messed up that fsck *refused to work*, not unless the disk was completely destroyed. That hasn't been a problem since ncheck and clri got replaced with fsck back in the '70s!

    And in three years, HFS+ ate its own brains three times... and that's just with my own personal hardware.

    That's worse than Windows.

    When they announced that Panther was getting an updated version of UFS from FreeBSD, I was hopeful. They *almost* had all the HFS+ extra functionality working over UFS in Jaguar... including aliases. But NO, they decided they were going to make the system even more HFS+-specific by putting the Spotlight hooks in the file system instead of the vnode layer.

    So until I actually see this come out from Apple, I'd take it with a grain of salt. They've had an appalling tendency to hold onto old bad ideas long past their freshness date.

    1. Re:If Apple wanted a better file system... by Anonymous Coward · · Score: 0

      UFS partitions can't be shrunk. If you get UFS, the say good-bye to Boot Camp, unless you want to do a fresh install of OS X.

    2. Re:If Apple wanted a better file system... by OrangeTide · · Score: 1

      HFS+ is technically superior in design to UFS. even if the utilities for it (like fsck) suck and it's far less reliable than UFs. Apple likes to cling to ideas that look on paper, even if they need a lot more work. ZFS is of interest because not only is it feature-packed, it apparently works too. It also is the right time to move away from HFS+ now that resource forks are almost never used, and there are reasonable ways to emulate them that work even on old UFS.

      If you're running a powerpc mac though, sticking with at least one HFS+ partition is still in your best interest. since the ROMs can't really boot OSX on much else and it is unlikely Apple will put any effort into making ZFS run native on the PowerPC port. Even if Linux/yaboot might support ZFS root on powerpc mac one day. it is possible to do with a trampoline bootloader, which is how yaboot manages to boot ext3/reiserfs/xfs partitions on a ROM that only understand HFS/HFS+, UDF, ISO and some variants of UFS.

      --
      “Common sense is not so common.” — Voltaire
    3. Re:If Apple wanted a better file system... by argent · · Score: 1

      HFS+ is technically superior in design to UFS.

      HFS+ is a different design. It has advantages in some areas, disadvantages in others, and a lot of those advantages *are* purely theoretical... for example the spotlight features can be implemented on top of *any* file system if they're handled at the vnode layer, and SoftUpdates shows that most of the supposed advantages of other file system structures are simply a matter of write scheduling and disk allocation algorithms that are applicable to any file system. So while there are advantages, they're fewer than you'd think.

      With those advantages come a lot of complexity, and attendant fragility. It has *never* been as reliable as UFS, and it's *not* just the utilities. The catalog can be completely corrupted just by running the disk near capacity, and that's *still* true in Tiger.

      You're right, Apple likes to cling to ideas that look good on paper. It's burned them over an over again, from the cooperative multitasking of the original OS, through the Copland fiasco, and even the use of Mach... OS X by now uses very little of Mach's features... it's just a performance sucker. And they have a terrible "Not Invented Here" problem. I'm afraid that no matter how good ZFS is, Apple will keep clinging to HFS+ no matter what.

    4. Re:If Apple wanted a better file system... by argent · · Score: 1

      If you get UFS, the say good-bye to Boot Camp, unless you want to do a fresh install of OS X.

      Imprimus, if I wanted to run Windows on my Macbook I wouldn't have bought a Macbook, I'd have bought a better and cheaper laptop with a keyboard and trackpad that didn't aggravate my RSI.

      Secundus, I'm going to have to do a fresh install of OSX pretty soon anyway. I don't know if it's HFS+ corruption, prebinding corruption, or what, but if I can't track down what's causing those occasional panics I'm getting it's going to be back to backups pretty soon.

      Tertius, for the majority of the switchers who think they need Boot Camp (Parallels or VMWare gives a MUCH better solution for everything but gaming) a reinstall of their new laptop or desktop is hardly a hardship.

      Quartus, I'm sure Apple will be able to find some features that ZFS doesn't do as well as HFS+, and THAT is the real point. Apple's reluctance to get rid of horribly bad ideas is legendary. I won't believe they're really switching until I see them switch... I've been burned too often by "Apple is finally coming to their senses" rumors too often. And even when my pessimisim *isn't* justified, they've managed to screw things up all too often. Apple's real advantage is that they don't actually *hate* their customers the way Microsoft does, or secretly despise them as the Linux community does. They're just incurably conservative beneath their progressive front...

    5. Re:If Apple wanted a better file system... by OrangeTide · · Score: 1

      I am hoping the hype around ZFS is enough to get Apple's attention. It would especially be interesting for those clustered Xserves talking over SAN. And just having "the best" file system around even if it wasn't invented by Apple seems like the sort of thing Steve Jobs would go for.

      --
      “Common sense is not so common.” — Voltaire
  60. Paths are evil by Rosyna · · Score: 1

    On a case-insensitive filesystem, your done if you're lucky. If not lucky, you need to do a linear scan of the whole damn directory. Many places have a directory with some insane amount of files. Intentionally or not, it's common to go into the tens of thousands. A few places (running XFS mainly, sometimes Reiserfs) get into the millions.

    And if you don't use paths, none of this applies at all. Paths are evil, slow, and fragile. Abstracted references, ftw! One for non-persistent data, one for persistent data. Yayness.

  61. "No" on all three counts by LKM · · Score: 2, Informative

    No, no and no.

    • ZFS will have case-insensitive support if it is used in Mac OS X (closed approved fast-track 05/09/2007).
    • HFS+ has case-sensitity as an option, so if you want to, you can have a case-sensitive Mac.
    • Case-sensitivity is a stupid idea for regular users and will never be on by default on Macs (there's no reason why "My Letter to Aunt Emma" should be a different file from "My letter to Aunt Emma")