Slashdot Mirror


Best Format For OS X and Linux HDD?

dogmatixpsych writes "I work in a neuroimaging laboratory. We mainly use OS X but we have computers running Linux and we have colleagues using Linux. Some of the work we do with Magnetic Resonance Images produces files that are upwards of 80GB. Due to HIPAA constraints, IT differences between departments, and the size of files we create, storage on local and portable media is the best option for transporting images between laboratories. What disk file system do Slashdot readers recommend for our external HDDs so that we can readily read and write to them using OS X and Linux? My default is to use HFS+ without journaling but I'm looking to see if there are better suggestions that are reliable, fast, and allow read/write access in OS X and Linux."

253 comments

  1. UFS. by necroplasm · · Score: 2, Informative

    UFS would be the best option. Linux supports it with -rw since Kernel 2.6.30 (afaik) and OS X mounts UFS natively.

    1. Re:UFS. by Moblaster · · Score: 1

      If you spend millions on an MRI machine, you should be able to ask the manufacturer to produce more reasonable software that splits those 80GB files into a number of smaller ones that most filesystems can manage.

      But why are you trying to roll your own solution anyway with such a proprietary system anyway?

    2. Re:UFS. by ecloud · · Score: 1

      Seems to be an oldie. Is it better than HFS? More reliable or higher performance? Still requires the occasional fsck right?

    3. Re:UFS. by clang_jangle · · Score: 4, Informative

      UFS would be the best option.

      Unless you're using Tiger or earlier, UFS is not an option. The last two versions do not support UFS at all. However, HFS+ support in Linux is pretty good. Otherwise you're looking at mac-fuse for ext2/3, which IME is pretty slow and buggy. I thinks Jobs has gone out of his way to make OS X incompatible with OSes other than windows. Maybe he's afraid of what will happen if everyone becomes aware they have other choices.

      --
      Caveat Utilitor
    4. Re:UFS. by kestasjk · · Score: 1

      It's the default filesystem in *BSD, so it's very well maintained etc. It has journalling (or does it call it "soft updates"?) auto-defrag, etc, etc. You fsck it if you power off without umount but otherwise you won't need to.

      It's definitely a perfectly capable, full-featured, modern filesystem.

      --
      // MD_Update(&m,buf,j);
    5. Re:UFS. by necroplasm · · Score: 1

      I wasn't aware of that. When I used UFS with Leopard it worked ok. It seems they crippled it in Snow Leopard, so it's of almost no real use.

      Too bad. Typical for Apple though. Such a shame.

    6. Re:UFS. by X0563511 · · Score: 2

      Every filesystem warrants the occasional check. If you never check, there are lots of errors that can accumulate and burn your ass.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    7. Re:UFS. by edman007 · · Score: 1

      Really? I remember I tried UFS around 10.4 and it was just a nightmare, every BSD seems to have their own version of UFS and linux could never autodetect it just right for me, I eventually gave up and went with HFS+ which worked just fine with the minor exception that unmounting HFS+ didn't set the flags right and I had to run some HFS+ app in linux (or boot OSX) to reset the flags so it would mount. I assume some of that has changed, but HFS+ still works in linux without a problem and the apple software likes it a bit better than UFS.

    8. Re:UFS. by ceoyoyo · · Score: 1

      MR scanners usually produce individual files that are smaller than a MB. I think the poster was referring to the total size of the dataset.

      It's quite possible that when they analyze the images they put them in a format where individual files are considerably larger though. It's a pain to do 3D, 4D or 5D analysis on a set of 2D files.

    9. Re:UFS. by Anonymous Coward · · Score: 0

      NFS/CIFS in the cloud; only say this as currently working on a cloud offering around this. Wondering how the pricing is going to work out for connections though.

    10. Re:UFS. by BlackBloq · · Score: 1

      Mac doesn't exist before tiger as a real beast so who fucking cares. IE they get what they deserve running osx9 and trying for interoperability. So really I would think they are not on Os9 or the answer should be "stop running old ass shit".

    11. Re:UFS. by Anonymous Coward · · Score: 0

      At first I thought you ment UDF. I have no idea how good UDF read/write support is in either MacOS X or Linux, which versions are supported, and what restrictions exist regarding filesize, filenames etc., but I think that UDF might just be the replacement for FAT I've been waiting for.

    12. Re:UFS. by Haxzaw · · Score: 1, Flamebait

      you were doing fine until you felt the need to bash on the OS. IMHO that was unnecessary and discredited your remarks. Whatever, I've come to expect that from the majority of /. Posters.

    13. Re:UFS. by RobertM1968 · · Score: 2, Funny

      ...storage on local and portable media is the best option for transporting images between laboratories. What disk file system do Slashdot readers recommend?

      Every filesystem warrants the occasional check. If you never check, there are lots of errors that can accumulate and burn your ass.

      Methinks you may be plugging your portable media into the wrong place... then again, I've never tried that, so I could be wrong. ;-)

    14. Re:UFS. by dogmatixpsych · · Score: 3, Informative

      Yes, the raw files from the scanner are quite small. A whole series of scans (7 or 8 high quality sequences) is only about 450MB. We get 80GB files when we do post-processing (fiber tracking) of a diffusion scan.

    15. Re:UFS. by Pharmboy · · Score: 1

      Which one? FAT12 and its 32MB limit? One of the FAT16 versions? FAT32 with 2GB limits? FATX? exFAT? TFAT? TexFAT?

      --
      Tequila: It's not just for breakfast anymore!
    16. Re:UFS. by Anonymous Coward · · Score: 4, Informative

      It's the default filesystem in *BSD, so it's very well maintained etc. It has journalling (or does it call it "soft updates"?) auto-defrag, etc, etc. You fsck it if you power off without umount but otherwise you won't need to.

      It's definitely a perfectly capable, full-featured, modern filesystem.

      All the things you write are perfectly true... on *BSD variants where UFS is the native, default FS. That is not the case on either Linux or OS X, to the extent that in OS X v10.6 UFS is now a read-only FS because it's barely maintained.

      Most people who think OS X is truly 'native' on UFS because it has BSD heritage haven't tried to actually use it. When Apple bought NeXT in 1997 the UFS implementation was already behind the times because at that time NeXT hadn't been updating its operating system for a few years. Since Apple wanted OS X to be a MacOS upgrade, development resources went into making a robust and high performance HFS+ implementation. Very little was done to modernize UFS. From the outside, it seems to have been just enough effort to make sure it worked and was still bootable over the first few versions, for those who wanted native UNIX FS semantics (mostly case sensitive file names). Then they added case sensitive filename support to HFS+ (it's a format-time option), and since then there has been even less reason for Apple to maintain UFS, hence its transition to a read-only legacy format.

      The other piece of this picture is that UFS != UFS. The UFS in MacOS X is a mildly upgraded version of mid-1990s NeXT UFS (which, in fine BSD tradition, wasn't quite the same as the UFS found in other BSDs). It's almost certain it has few of the features you associate with modern versions of UFS.

    17. Re:UFS. by quickOnTheUptake · · Score: 1

      Here. In my limited experience I seem to recall that Vista implemented some oddities that made it impossible to read a UDF disk made on Vista on a non-windows machine.

      --
      Mod points: Guaranteed to remove your sense of humor.
      Side effects may include gullibility and temporary retardation
    18. Re:UFS. by dezent · · Score: 0

      According to wikipedia. "Currently there is a 4GB file limit for disks formatted as UFS in Mac OSX." http://en.wikipedia.org/wiki/Unix_File_System

    19. Re:UFS. by Thinboy00 · · Score: 1

      Just because people regularly bash the company for dumb reasons doesn't mean that every time someone bashes the company it's for a dumb reason.

      ObTopic: Just wipe all the OSXen and replace them with Ubuntu 10.04. Install AWN and Compiz, configure keybindings and themes, and no one will know the difference. Then you can use ext4.

      --
      $ make available
    20. Re:UFS. by X0563511 · · Score: 1

      But I thought they were Plug and Play!

      (oh god, did I just say that?)

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    21. Re:UFS. by oakgrove · · Score: 1

      Wait a minute, where did he bash on any OS? Steve Jobs, maybe. But what he said was pretty tame. Hardly a bash.

      --
      The soylentnews experiment has been a dismal failure.
    22. Re:UFS. by Anonymous Coward · · Score: 0

      Actually, I thought the remarks about Jobs pretty much described the situation as is. Not to discredit all the contributions Steve Jobs has made to computing but ever since he came back from his life-threatening health problem he's been much more antagonistic and confrontational. Really, just what is so inaccurate about a summation the Apple is moving more and more into a 'closed garden' style of development?

    23. Re:UFS. by RobertM1968 · · Score: 1

      But I thought they were Plug and Play!

      (oh god, did I just say that?)

      LOL!!!!! Gives the phrase a whole new meaning... and I think I will be handling my customers' external hard drives (or should I have put "external hard drives" in quotes?) with gloves from now on... ;-)

    24. Re:UFS. by jordan_robot · · Score: 1

      FAT chicks -- they eat all your data & keep coming back for more.

    25. Re:UFS. by Anonymous Coward · · Score: 0

      Maybe he's afraid of what will happen if everyone becomes aware they have other choices.

      What are they going to do? Are they all going to run out and get an Ubuntu CD and install over their Windows and OS X installs?

      Right.

    26. Re:UFS. by BrokenHalo · · Score: 1

      ask the manufacturer to produce more reasonable software that splits those 80GB files into a number of smaller ones that most filesystems can manage.

      In my case, I use HFS+ on my external HDDs. Linux has supported it for some time, and I have had no problems with it. UFS has only recently become supported as -rw in Linux, and as such should (IMO) should be treated with some caution.

      I avoid any of the FATs or NTFS because I don't like the way they fuck up *nix permissions.

    27. Re:UFS. by toQDuj · · Score: 1

      Except that linux (in my experience) is only able to mount HFS+ read-only.

      --
      Every experiment which ends in a big bang is a good experiment.
    28. Re:UFS. by Wescotte · · Score: 1

      Does linux support HFS+ w/ journaling now?

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

      you need a pacs. Something like http://www.osirix-viewer.com/ since you state you have macs. But a pacs of some kind is what you need to best keep track of and protect patient information though access rights.

    30. Re:UFS. by spazdor · · Score: 1

      I've converted 6 non-geeks in the past year (4 of those since Lucid). Are you doing your part?

      --
      DRM: Terminator crops for your mind!
    31. Re:UFS. by dotgain · · Score: 2, Informative

      You can get R/W support but only if you disable the journal on the HFS+ fs on MacOS. That's unless they've taken that away too in 10.6, but I routinely do this with HFS+ between 10.5 and Linux.

    32. Re:UFS. by vegiVamp · · Score: 1

      Some portable harddisks vibrate quite a lot, you know.

      --
      What a depressingly stupid machine.
    33. Re:UFS. by aliquis · · Score: 1

      I don't know to what extent but surely some of the BSDs and also Solaris UFS are different. So it's not totally the same filesystem, though maybe similar enough.

    34. Re:UFS. by aliquis · · Score: 1

      Don't worry, mine uses NiMH.

    35. Re:UFS. by aliquis · · Score: 1

      Methinks you may be plugging your portable media into the wrong place..

      Also make this story much more hillarious:
      MS Design Lets You Put Batteries In Any Way You Want

    36. Re:UFS. by aliquis · · Score: 1

      Seeing the shapes of hard-drives being somewhat awkward for the imagined purpose I think you can spare the gloves for their joy-sticks, eventually use them for handling wii-wands to, though there are already replaceable rubber molds to cover those in.

    37. Re:UFS. by aliquis · · Score: 1

      Too bad we lost the ZFS option. Though maybe not willing to use fuse in Linux either, but for OS X atleast.

    38. Re:UFS. by aliquis · · Score: 1

      You would have had a point if he said UFS only worked in Leopard and Snow leopard but not in Tiger and earlier.

      However he sad the opposite... It did worked in "the old ass shit" but not in the new versions.

    39. Re:UFS. by Anonymous Coward · · Score: 0

      I do my best to convince them to stay of Apple products.

    40. Re:UFS. by superposed · · Score: 1

      "I thinks Jobs has gone out of his way to make OS X incompatible with OSes other than windows."

      I wouldn't say that OS X is particularly compatible with Windows either. There isn't any file system that can handle >2 GB files and is available for free (in either sense) on Windows and OS X. Windows uses NTFS but has no built-in support for HFS+; Mac uses HFS+, but has only read-only support for NTFS. Neither of them will read or write ext2/ext3, and neither of them has good support for UFS.

      Linux gets good marks for supporting all of the above, but Apple and Microsoft can't seem to get their act together. This is probably because Microsoft has never seen fit to support HFS+ (which is publicly documented, although I don't know if it's an open standard) or to publish the specifications for NTFS (which would allow Apple to support it natively).

      If you're willing to install third-party software, the options open up a bit. MacDrive does a good job of accessing HFS+ from Windows and HFSExplorer or Apple's Boot Camp driver gives read-only support. MacFUSE can access NTFS and ext2/3, but was slow and/or unstable when I last tried it.

      I'm amazed that in 2010 there's still no widely supported modern file system for all three operating systems. It's a big pain if you want to run a mix of OS X, Boot Camp and Parallels on the same machine -- there's no way to keep all your documents in one place.

    41. Re:UFS. by Anonymous Coward · · Score: 0

      ObTopic: Just wipe all the OSXen and replace them with Ubuntu 10.04. Install AWN and Compiz, configure keybindings and themes, and no one will know the difference. Then you can use ext4.

      The main thing your comment proves is that you've never used a Mac. There is no X-window manager/DE that comes anywhere close to being "Mac-like". Seriously. The UI is a completely different paradigm. Not to mention that all those gtk/qt apps are jarringly grotesque compared to aqua apps. No Mac user would ever be fooled by an X-windows setup.

      It gets old, reading all the criticisms of OS X from Linux fanbois who are obviously inexperienced with OS X...

    42. Re:UFS. by Anonymous Coward · · Score: 0

      Then either you or your distro of choice are doing it wrong. Linux has had flawless rw support for HFS+ since at least 2006 now.

    43. Re:UFS. by clang_jangle · · Score: 1

      you were doing fine until you felt the need to bash on the OS. IMHO that was unnecessary and discredited your remarks.

      I didn't "bash" anything or anyone, just calling it as I see it. If you think about the fact that it would be trivial for Apple to support as many filesystems as Linux does, it becomes obvious that what I said has merit. Steve Jobs may have been Apple's saviour as far as profitability goes, but he's also been responsible for making some kinds of interoperability very difficult. Lack of support for filesystems other than FAT and HFS+ is just one example, but IMHO it's inexcusable. With a userland rooted in BSD there's just no good reason why OS X should not be the king of interoperability, it would be so easy! But the same kind of thinking that causes a Linux or BSD install on a Mac to be labeled "Windows" in the boot menu (because "Mac" and "Windows" are the only OSes acknowledged by the boot code) means you can't have any filesystem Jobs doesn't like. Only the most hardcore of fanbois would not see the problem with this.

      A protective, walled garden for the iPhone is one thing (though not my thing, BTW) -- withholding what should be basic functionality on a general-pupose computer is something else altogether. I bought my first computer, a Mac SE, in 1987. My current Mac, purchased in 2009, will probably be my last. But I suppose the millions of new customers gained selling CE devices will more than compensate Apple for the few serious geeks who will migrate to other brands. They seem to have lots of rabid non-geek fans now.

      --
      Caveat Utilitor
    44. Re:UFS. by kristjansson · · Score: 1

      no one will notice the difference, until they, you know, try to do the things that they've been hired to do, most likely. I don't think that they have macs alongside linux boxen just for grins... and in any case, those machines may not be his to do with as he pleases, to begin with, since he's talking about shuttling data for clients, not merely client sites...

    45. Re:UFS. by sh00z · · Score: 1

      MacDrive does a good job of accessing HFS+ from Windows and HFSExplorer or Apple's Boot Camp driver gives read-only support. MacFUSE can access NTFS and ext2/3, but was slow and/or unstable when I last tried it.

      NTFS-3G offers excellent (and free as in beer) NTFS read/write support all the way back to OS 10.4 PPC. I've found it very stable.

    46. Re:UFS. by Thinboy00 · · Score: 1

      There's this grand new invention, it's called humor.

      --
      $ make available
  2. Followup question... by serviscope_minor · · Score: 2, Informative

    I have a similar problem, albeit on a smaller scale. I use unjournalled HFS+.

    However, the problem is that HFS+ being a proper unix filesystem remembers UIDs and GIDs which are usually inappropriate when the disk is moved.

    Is there any good way to get Linux to mount the filesystem and give every file the same UID and GID, like for non unix filesystems?

    --
    SJW n. One who posts facts.
    1. Re:Followup question... by SEAL · · Score: 2, Informative

      Many filesystems support uid= and gid= options in their mount command (including HFS). Just add that to a mount script or set it up in fstab.

    2. Re:Followup question... by X0563511 · · Score: 4, Informative

      Non-native filesystems usually let you set UID, GID, and permission masks. Check the "mount" manpage and look for the filesystem you want. You might also try "man filesystem"

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    3. Re:Followup question... by Anonymous Coward · · Score: 0

      You give this guy an informative for "man filesystem". I've been gone a while but I had no idea.

    4. Re:Followup question... by Thinboy00 · · Score: 1

      On my computer, at least, I think he meant "man filesystems" (plural) since man filesystem is about the signaling event, which isn't very helpful at all.

      --
      $ make available
    5. Re:Followup question... by Jon_E · · Score: 1

      of course you could also just cross-mount NFS, or setup a central networked fileserver .. as network begins to surpass native FC speeds - it's not a bad option (but needs a bit of tuning, care and feeding)

      I played a bit with zfs-fuse too for doing this sort of thing (along with native zfs on opensolaris), but with a stall in cross platform development here i ended up upgrading my pools and obsoleting the fuse versions .. (note: also a bit tricky given the panic on detach model)

    6. Re:Followup question... by inKubus · · Score: 1

      Just have a unique UID for each user in the company.. Start at 10000. Store in LDAP using posixuser schema. Use nsswitch to direct /etc/passwd to ldap. Problem solved.

      --
      Cool! Amazing Toys.
    7. Re:Followup question... by Willfon · · Score: 1

      sudo diskutil disableOwnership /Volumes/MovableDiskName
      I think that toggles a bit in the filesystem itself, making it ID unaware. But I haven't tried to see how it looks on Ubuntu afterwards. Go go gadget VirtualBox!

      --
      kwik-mart
    8. Re:Followup question... by Anonymous Coward · · Score: 0

      Why don't you use the same UID for your account everywhere?

    9. Re:Followup question... by PolarIced · · Score: 1

      On removable devices you can also click the device in the Finder and select "Get Info." There is a checkbox for "Ignore Permissions on this Volume." This is more Mac to Mac, but it might help if you're going back and forth between several machines. Just thought I'd point it out.
       

  3. NTFS by Anonymous Coward · · Score: 0

    NTFS :)

    1. Re:NTFS by X0563511 · · Score: 3, Funny

      Windows doesn't play in here, it's OSX and Linux. Tossing NTFS into that would just be... wrong somehow.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    2. Re:NTFS by Tirs · · Score: 5, Funny

      Hah! In my company we call it "NoTeFíeS" (for non Spanish-speaking people: "Don'tTrustIt").

      --
      Strength, balance, courage and reason. If you know what's this about, contact me!
    3. Re:NTFS by nut · · Score: 1

      NTFS is actually not a bad option. Ubuntu 10.04 supports it out of the box (sic), OS X supports read by default but not write. Several companies supply cheap RW drivers for NTFS on OS X.

      --
      Never trust a man in a blue trench coat, Never drive a car when you're dead
    4. Re:NTFS by legrimpeur · · Score: 1

      agreed I have been using over FAT as it supports files larger than 2GiB. Through FUSE you can get RW capabilities for free http://www.tuxera.com/community/ntfs-3g-download/ . A bit slow in writing though...

    5. Re:NTFS by Anonymous Coward · · Score: 0

      Exactly. One of the systems should have native support for the fs at least .. right?

    6. Re:NTFS by MachineShedFred · · Score: 1

      with 10.6, the supplied NTFS driver can do read / write, but it's not supported by Apple. You just need to change it to RW in /etc/fstab.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    7. Re:NTFS by FreonTrip · · Score: 4, Funny

      Roses are red,

      Violets are blue,

      All of my mod points

      Would belong to you.

    8. Re:NTFS by cyrano.mac · · Score: 1

      Quote from fstab.hd: "IGNORE THIS FILE. This file does nothing, contains no useful data, and might go away in future releases. Do not depend on this file or its contents."

    9. Re:NTFS by Irontail · · Score: 1

      I'd also suggest NTFS. Both OSes support reading and writing it. In my experience, ext2/3/4 support in MacFUSE is crap, at least since Macs went Intel. NTFS also has the advantage of supporting files over 4GB, unlike something old like FAT32.

      NTFS is kinda counter-intuitive, yeah, but it works. *shrug*

    10. Re:NTFS by gabebear · · Score: 1
      In practice it usually works fine, but I have seen issues when a drive is used between XP and Vista. NTFS is undocumented and has wacky features(like ActiveX NSS).

      From Wikipedia:

      While the different NTFS versions are for the most part fully forward- and backward-compatible, there are technical considerations for mounting newer NTFS volumes in older versions of Microsoft Windows. This affects dual-booting, and external portable hard drives.

      For example, attempting to use an NTFS partition with "Previous Versions" (a.k.a. Volume Shadow Copy) on an operating system that doesn't support it, will result in the contents of those previous versions being lost.

      - http://en.wikipedia.org/wiki/NTFS

    11. Re:NTFS by RobertM1968 · · Score: 3, Informative

      Windows doesn't play in here, it's OSX and Linux. Tossing NTFS into that would just be... wrong somehow.

      Flamebait mod or not, there is a valid point. Though various NTFS drivers do allow read/write, the success isn't graven in stone. There are better alternatives in the Linux/OSX world. Keep in mind that losing this data becomes either costly (as in time=money, let's go make another set of copies to run to whatever office) or very bad (as in someone moved the files to the external instead of copying them) or both.

      So, as good as the NTFS R/W drivers are getting, it's safer to use a file system that is known to be more stable and less error prone, such as HFS+ or UFS or one of the other suggestions. "Really good" shouldnt be an option in the medical world when "even better than 'really good'" is available, compatible, and easy to install on all systems involved.

    12. Re:NTFS by Anonymous Coward · · Score: 0

      NTFS is actually not a bad option. Ubuntu 10.04 supports it out of the box (sic), OS X supports read by default but not write. Several companies supply cheap RW drivers for NTFS on OS X.

      Unfortunately, I've found NTFS to be the only real option for sharing disks between Linux and Mac. Oh, and BTW, you don't have to pay for decent NTFS support on OS X: http://macntfs-3g.blogspot.com/ (grab the ones that say ntfs-3g)

    13. Re:NTFS by dakameleon · · Score: 1

      Have you got any source to confirm this or are you just pulling it out of your ass? Why would that not have been made wider knowledge?

      --
      Man who leaps off cliff jumps to conclusion.
    14. Re:NTFS by Anonymous Coward · · Score: 0

      Mod parent up here, I would but I can't even log in from the office...
      NTFS-3g works GREAT when you're accessing your windows drive...most of the time. I use it all the time helping out friends with their machines. It is, however, somewhat unstable when it comes to writing to the NTFS file system, especially for large writes like the OP is talking about. Honestly either ext2 or hfs+ would be the way to go. Having not used the ext2 tools for mac, I can't comment on their performance however, so I can't make further comparison between the two.

    15. Re:NTFS by Thinboy00 · · Score: 1

      Quote from fstab.hd:

      "IGNORE THIS FILE.
      This file does nothing, contains no useful data, and might go away in
      future releases. Do not depend on this file or its contents."

      ...which leaves me wondering, WTF?

      Is the rest of the file there like in Linux? Is there a "rest of the file"?

      --
      $ make available
    16. Re:NTFS by Anonymous Coward · · Score: 0
      How the hell is this flamebait? You M$ shill mod astroturfers need to go get fucked somewhere. How is somebody in a professional capacity supposed to use a reverse engineered patent encumbered filesystem in a non-native fashion on 2 different OS's that it wasn't even designed to run on in the first place? Not to mention the performance of fuse is woefully inadequate compared to anything native. Everytime you fracking write to it, you'd better cross your fingers that all of the bits actually make it to the platter.

      Tossing NTFS into that would just be... wrong somehow.

      I'll drink to that.

    17. Re:NTFS by Cramer · · Score: 1

      NTFS is the worst option. It was NEVER intended to be a removable filesystem, even less a portable filesystem. For drives that only get used on the same computer, it'll work. However, the instant you transport any drive to other computers, you will start having problems from ACLs -- if you're using XP Home, you have no simple way to fix it. This is esp. true of systems that aren't part of the same domain since every system will have a unique set of users.

    18. Re:NTFS by X0563511 · · Score: 1

      Woa woa, calm down. I don't need an AC freaking out on my behalf :)

      I can handle the karma hit. On this post too.

      (thanks though)

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    19. Re:NTFS by zaphod777 · · Score: 1, Informative

      easy enough to solve just propagate "everyone" read write access to the drive.

      --
      "Don't Panic!"
    20. Re:NTFS by catmistake · · Score: 1

      It's complicated. Using NTFS without Windows seems like it would be less complicated, doesn't it? One would need to perform all sorts of tests paying close attention to resource forks and file names. And then even if you're not using the platform, check it in Windows... I can't think of a good reason why, other than Windows will bite you in the backside when you least expect it.

      Recently I've been having some issues with Case-Sensitive Journaled HFS+ and I'm now looking for the ideal FS for the replacement NAS, but I don't have the constraints of the OP.

    21. Re:NTFS by yuhong · · Score: 1

      http://www.macosxhints.com/article.php?story=20090913140023382 Comments say that it is not very stable, though

    22. Re:NTFS by Anonymous Coward · · Score: 0

      This is fantastic - I have just downloaded and installed on my MAC. Sweeet. I always thought NTFS locked out because of patents and the propriatary nature of the NTFS file system (TomTom got sued just for having a system that converted long filenames to the same 8.3 character names MS use on a FAT32 disk after all)

    23. Re:NTFS by Kjella · · Score: 1

      Nothing is guaranteed in stone, but if there's a threat it's that Microsoft has implemented something odd which the driver doesn't handle and fails spectacularly on. I haven't looked at exactly how NTFS driver support has been done, but if you're running the same ntfs-3g driver on Linux and Mac (it is Unixy enough for that, isn't it?) without touching a Windows box, you should be even safer. Not that I've had any problems with it for years.

      --
      Live today, because you never know what tomorrow brings
    24. Re:NTFS by RobertM1968 · · Score: 1

      Nothing is guaranteed in stone, but if there's a threat it's that Microsoft has implemented something odd which the driver doesn't handle and fails spectacularly on. I haven't looked at exactly how NTFS driver support has been done, but if you're running the same ntfs-3g driver on Linux and Mac (it is Unixy enough for that, isn't it?) without touching a Windows box, you should be even safer. Not that I've had any problems with it for years.

      Very true... but most of the medical type establishments I have worked in have a plethora of Windows machines that the data gets transferred to as well... so even with various of the special purpose machines running MacOSX or Linux, there's a really good chance that the stuff will end up being transferred to (and/or from) a Windows machine. For instance, being printed out by a nurse or an assistant from their Windows workstations, with notes or timestamps or logs (of usage) being updated.

    25. Re:NTFS by gabebear · · Score: 1

      Why are you needing case-sensitivity? I've run into several Mac programs that expect to be on case-insensitive storage systems.

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

      lol spic

    27. Re:NTFS by MachineShedFred · · Score: 1

      I didn't say fstab.hd

      I said /etc/fstab which does do something. It allows you to enable write access to NTFS volumes.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    28. Re:NTFS by Anonymous Coward · · Score: 0

      Those applications are, then, quite old and in need of updating. I think CS2 didn't do Case-sensitive... so I installed it on it's own case-insensitive disk image, and it works fine. But Adobe did eventually update the software to be compatible, in case you walked out of the theater like a lot did and missed the end of that movie, it did actually happen, and such a surprise, too.

  4. Answers Own Question. by Rantastic · · Score: 0

    My default is to use HFS+ without journaling but I'm looking to see if there are better suggestions that are reliable, fast, and allow read/write access in OS X and Linux.

    So I see you're the "if it ain't broke fix it anyway" type.

    --
    Ask Slashdot: Where bad ideas meet poor googling skills.
    1. Re:Answers Own Question. by octothorpe99 · · Score: 0

      That statement does imply that the OP finds HFS+ to be either "unreliable" or "slow" or both. So it may in his opinion be "broken"

    2. Re:Answers Own Question. by Anonymous Coward · · Score: 0

      My default is to use HFS+ without journaling but I'm looking to see if there are better suggestions that are reliable, fast, and allow read/write access in OS X and Linux.

      So I see you're the "if it ain't broke fix it anyway" type.

      Or the "it may not be broke, but I might be able to make it better" type. You know, like the sort that tends to actually make things better from time to time?

      Just because it works doesn't mean it can't improve.

    3. Re:Answers Own Question. by Anonymous Coward · · Score: 0

      TBH, I really don't see what choice you have beyond HFS; OSX is your limitation. Whatever filesystems it supports, is what you have to use.

    4. Re:Answers Own Question. by Anonymous Coward · · Score: 0

      So I see you're the "if it ain't broke fix it anyway" type.

      Or the "It ain't broke, but is there a better way?" type- or are you still driving a horse and buggy around?

  5. Ugly but works by mewsenews · · Score: 0

    You will want to use vanilla FAT32 but then beef up the data with par2 and/or sfv files for error recovery and checksumming respectively.

    FAT32 isn't reliable as anything but a lowest common denominator between platforms, anything else will give you headaches especially if you need to share the drives outside your local ecosystem.

    QuickPAR and QuickSFV are the Windows utilities I've used, there are probably versions available on all platforms.

    1. Re:Ugly but works by GooDieZ · · Score: 2, Informative

      and 4Gb cap...

      --
      Things in a rear mirror might be behind you
    2. Re:Ugly but works by X0563511 · · Score: 1

      Indeed. You completely missed the 80gb file part.

      Do *nix and OSX support exfat at all? If they do, then that -should- work. But it's not really a good solution.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
  6. HIPAA Constraints? by fm6 · · Score: 5, Interesting

    By "HIPAA Constraints" I assume you mean the privacy rule. I would think that this rule would prevent you from using sneakernet to transmit files. Unless you're encrypting your portable disks, and somehow it doesn't sound like you are.

    Fun reading:

    http://www.computerworld.com/s/article/9141172/Health_Net_says_1.5M_medical_records_lost_in_data_breach

    1. Re:HIPAA Constraints? by ducomputergeek · · Score: 1

      That was my first thought as well. And as much as I hate to say it, but Fat32 might be the best option. Either that or UFS.

      --
      "The problem with socialism is eventually you run out of other people's money" - Thatcher.
    2. Re:HIPAA Constraints? by Anonymous Coward · · Score: 0

      except for the problem that fat32 can only handle files smaller than 4gb

    3. Re:HIPAA Constraints? by Anonymous Coward · · Score: 1, Informative

      Why would "sneakernet" be disallowed for digital medical files? This procedure is no different than transferring real physical medical files or records.

    4. Re:HIPAA Constraints? by Anonymous Coward · · Score: 0

      FAT32 has a 4GB file size limit and fragments too much, he needs something that can cope with huge 80GB files.

    5. Re:HIPAA Constraints? by hedwards · · Score: 1

      NTFS is probably the best bet. I don't like saying that, but it is the most widely available filesystem that can handle the large size of files. But really the company that makes the equipment really ought to come up with some way of reducing the size of the files to something reasonable. Most likely via splitting of some sort.

    6. Re:HIPAA Constraints? by X0563511 · · Score: 1

      But.... but it involves COMPUTERS! It's completely different, we need new rules!

      (that said, it's far easier to pocket a USB drive (or just copy) and run then a folder full of files or some x-ray prints)

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    7. Re:HIPAA Constraints? by cynyr · · Score: 1

      I'm pretty sure MRI data is that size for a reason...

      --
      All of the above was encrypted with a Quad ROT-13 method. Unauthorized decryption is in violation of the DMCA.
    8. Re:HIPAA Constraints? by eschasi · · Score: 4, Informative
      HIPPA mandates who can and should have access to the files. The method of storage (disk, tape, SSD, paper, whatever) is largely irrelevant. As long as all those who have access to the files are HIPPA-trained and following the appropriate procedures, everything is fine. Similarly, transport is relevant only in that there must be no data disclosure to unauthorized persons. As such, if a person with appropriate clearance does the transport, all is cool.

      HIPPA data is often encrypted when placed on tape or transported across systems, but that's because such activities may involve the data being visible to unauthorized people. As examples of each:

      • If two physically separate sites exchange HIPPA data across the open Internet, the data must be encrypted during transport. This might be done by VPN, sftp, whatever. As long as the bits on the wire can't be read by the ISPs managing the connection, it's OK.
      • For tapes that you archive off-site, you don't want your external storage facility to be able to read the tapes, nor have the data readable if the tape is misplaced in transport.

      IMHO wise use of sensitive data on laptops requires encryption at the filesystem level. It's neither difficult or time-consuming, but given how much sensitive data has been exposed via folks losing or misusing laptops, it ought to be a no-brainer. Sadly, too few places bother.

    9. Re:HIPAA Constraints? by fm6 · · Score: 1

      Oh yeah, don't put it on one 80gb disk drive, put it on forty 2gb thumb drives. Nothing can go wrong with that.

      If these guys are going to be transferring a lot of 80gb files, they have to find a way to do it over the network securely and reliably. Not easy, I admit, but using sneakernet for this kind of data (even without the huge file sizes) is asking for trouble.

    10. Re:HIPAA Constraints? by fm6 · · Score: 1

      FAT32 was pretty fat when it came out 15 years ago. Nobody even had 2GB drives, never mind 2GB files.

    11. Re:HIPAA Constraints? by Anonymous Coward · · Score: 0

      Why would what *you* think matter at all when it comes to interpreting a complex set of requirements and rules. Have you read them? Have you been trained on them? Do you have to comply with them at your job? If not, why are you even commenting on them. Your idle speculation is worthless.

      A research institution is going to have an entire department and program devoted to HIPAA compliance, complete with training program. You obviously have not the first clue about what the rules mean or how to implement them, so please just concentrate on things you know about, which I'm guessing is a very manageable task.

    12. Re:HIPAA Constraints? by rwa2 · · Score: 2, Interesting

      Maybe instead of using a portable disk, they could whip up a nettop running Linux and transfer files over the gigabit ethernet...
      Then they could do transfers via samba or rsync+ssh , and the nettop could transparently take care of encrypting the underlying FS, whatever that may be.

      Performance wouldn't be great... maybe 20MB/s instead of 60MB/s for an eSATA drive, and they'd have to work out a consistent network port / IP across all the sites it travels to. But it might confer some advantages.

      Along similar lines, they could put in a small file server at each site, and rotate a removable disk drive between all of the file servers. That way they'd just have a drop box that they could always push files to throughout the day, and let the couriers just grab what's there and deliver.

    13. Re:HIPAA Constraints? by Leebert · · Score: 1

      Even though you speak as someone knowledgeable and authoritative about HIPAA, I have a hard time believing you since you apparently don't know how to spell it.

    14. Re:HIPAA Constraints? by dugjohnson · · Score: 1

      Absolutely correct. If you are putting the files on a USB drive without encrypting you are setting yourself up to become a news story, and not a good one. I'm going to assume that you are not in a facility that is planning on getting any of the incentive money from ARRA because any time you move outside the network (copying to USB, CD, DVD, another computer not constrained by the network) you HAVE to encrypt....you should anyway, but to qualify for meaningful use you must. If you have lots of files to share all the time, you might consider getting a PACS; there are open source PACS that have a lot of capability and would let you share, notate, and track access (HIPAA again). Take a look at dcm4chee

      --
      My brain is overly lubricated
    15. Re:HIPAA Constraints? by RobertM1968 · · Score: 1

      By "HIPAA Constraints" I assume you mean the privacy rule. I would think that this rule would prevent you from using sneakernet to transmit files. Unless you're encrypting your portable disks, and somehow it doesn't sound like you are.

      Fun reading:

      http://www.computerworld.com/s/article/9141172/Health_Net_says_1.5M_medical_records_lost_in_data_breach

      You would be surprised at how outdated parts of HIPAA are (from the day they were written). And what things they fail to cover. Heck, there are sections that indicate the requirement for data encryption for certain uses/storage/etc, but that's about the extent. ANY encryption will do to pass muster. A simple subsitution key would pass the required criteria. Then there are sections that are very specific in specifying methods that are useless... while others at least seem to have been thought out. There are sections that either were never written or not included in the final that should have been as well.

      It really seems like they hired no one of any security knowledge to write it. Oh, btw, I deal with this stuff for various apps we write... I was appalled at some of it...

    16. Re:HIPAA Constraints? by dogmatixpsych · · Score: 2, Interesting

      All of our scans are natively anonymized: we make up the birthdate and we never include the research participants' names. Our images are high enough quality that you can do nice 3D reconstructions of people's heads (and faces) but there is virtually no chance that anyone would recognize the faces unless they knew a person really well (even then it is hard). We have checked with the privacy office and our drives do not have to be encrypted because the images that will be on these drives are de-identified. We probably will encrypt the drives but wanted to nail down a filesystem first before we deal with encryption.

      Anyway, we only rarely will need to use the sneakernet but need the option. HIPAA is only a minor issue. The biggest issue comes in dealing with multiple IT departments and setting up network access to our materials. Plus our images are so large that for these processed files (not the originals) we are opting for local storage instead of storage managed by our IT staff (who are wonderful but not cheap; we just purchased 4TB of local storage for 1/4 the cost of 1TB from IT).

    17. Re:HIPAA Constraints? by dogmatixpsych · · Score: 1

      FAT32 has a 4GB file size limit. We have 80GB+ files (and could, if we wanted, have 250GB files but RAM becomes a limiting factor).

    18. Re:HIPAA Constraints? by dogmatixpsych · · Score: 1

      We would if it was easy but alas we have to deal with multiple IT departments in order to do that plus a lot of other red tape. The biggest issue is the file sizes (we're on gigabit but our colleagues are not), otherwise we would not use portable drives (well, money is a factor too; government grants only provide a limited amount of money).

    19. Re:HIPAA Constraints? by RobertM1968 · · Score: 1

      Even though you speak as someone knowledgeable and authoritative about HIPAA, I have a hard time believing you since you apparently don't know how to spell it.

      Well, as someone who is knowledgeable on it, he's pretty much right. But the sad part is, any encryption suffices to be HIPAA compliant. I've run into some pretty lame ass setups where such data was being stored on ancient Windows Server machines behind a "firewall" that qualified as meeting HIPAA requirements. The whole setup probably did (the part I saw did) - but, in realistic terms, was still highly insecure.

      HIPAA seems to be part "let's make an attempt - it doesnt matter if it's a good one" and part "this should appease the non-technical public, who wont know any better"

      Kinda like back in the day when the credit card companies, to ease credit card holders' minds, told people to "make sure you get the carbon!!! that'll protect you from credit card number theft!!!" - Ummm... duh... if anyone wanted to steal someone's credit card number back then (in the days of the imprint machine and carbon multi-"page" receipts), they'd simply write down the number off the very legible store or credit card company copies instead of trying to "decipher" it off the carbon sheets.

    20. Re:HIPAA Constraints? by fm6 · · Score: 1

      The biggest issue comes in dealing with multiple IT departments and setting up network access to our materials. Plus our images are so large that for these processed files (not the originals) we are opting for local storage instead of storage managed by our IT staff (who are wonderful but not cheap; we just purchased 4TB of local storage for 1/4 the cost of 1TB from IT).

      Dude, there's a reason network storage is more expensive than local storage: it comes with the infrastructure that allows lots of people to access it. If you try to serve up these large files from your local network, you'll slashdot the thing, and wackiness will ensue.

      Getting back to the privacy issue: I hope your privacy officer did due diligence, and isn't some overworked functionary who just said, "The data is anonymized? Well, that's OK then." You wouldn't be the first people to distribute data they thought was sufficiently anonymized, only to find that some clever data miner managed to connect the names with the data.

      http://www.wired.com/politics/security/commentary/securitymatters/2007/12/securitymatters_1213

    21. Re:HIPAA Constraints? by fm6 · · Score: 1

      HIPPA mandates who can and should have access to the files. The method of storage (disk, tape, SSD, paper, whatever) is largely irrelevant.

      Say what? You've never hear of a data breaches from lost or stolen portable hardware? See the link in the post you replied to.

    22. Re:HIPAA Constraints? by drfreak · · Score: 1

      I use TrueCrypt to transport patient data to/from doctor's offices.

    23. Re:HIPAA Constraints? by DigiShaman · · Score: 1

      And there's nothing wrong with external storage so long as the rules are being followed. Geologists transport huge SEG-Y files on 1TB drives these days. It's high latency, but very high bandwidth.

      --
      Life is not for the lazy.
    24. Re:HIPAA Constraints? by kramulous · · Score: 1

      I completely understand the red tape.

      Our scientists have been having similar problems. I believe that the real solution here is to stop these guys from working on their local machines with the full sized datasets. We've provided a centralised HPC system that is connected via infiniband (and others) to multiple architectures of storage.

      There is the standard /home which is DMF'ed with the top tier being 50T of total 650MB/s write (not sure of the read stat - I'm the software guy not the hardware guy). This is fine for a lot of people. They ssh into an interactive batch compute node for testing the datasets too big for a workstation (definitely for the Standard Operating Environment) and run their programs there. They can have access to up to 192GB of shared memory (12 nodes with that). At the moment, each researcher can have a maximum of 130 jobs running at any given time.

      There is also a /scratch that is made up of multiple units each being 32TB and individually capable of 750MB/s write - 600MB (single client - 800MB/s total for multiple reads). This is for the terabyte dataset analysis - your 80GB datasets are that for now ... they will grow quickly.

      These disks can be mounted to multiple areas around the country. Not all, given some are behind with getting their security certificates working properly. The problem with removable storage is that the researcher copies their dataset onto the drive and then deletes the original. One copy in a volatile situation.

      The trick here is to educate the user that working on their dataset in front of them is not sustainable. The data acquisition rate is much higher than Moore's law. Whether that is you have a 'cloud' like arrangement for those who still use standard Windows programs or a centralised pool of resources to some sort of cluster arrangement.

      --
      .
    25. Re:HIPAA Constraints? by dogmatixpsych · · Score: 1

      There is no way for names to be connected with the data without someone hacking our managed filesystem (it's possible, but that is IT's responsibility, not ours) or having direct access to our lab space (but if we are not there they would have to get through multiple locked doors and break into a locked cabinet).

      We opted to go with local, portable storage because only 4 people need or have access to these particular image files on three computers (we have 2 more collaborators that might need access but we can get them the raw data and they can recreate what we did on their computers). We would love to have the managed storage but it our grant only has so much funding so we want the most bang for our (your - assuming you live in the U.S. and pay taxes) buck.

    26. Re:HIPAA Constraints? by dogmatixpsych · · Score: 1

      We would love to have that infrastructure at our institution; we're working that way but we don't have it yet. If our lab could afford it we'd do something like that but our funding isn't that high. The other issue is that we use Macs because they are the best for neuroimaging work (Linux is great but things work better in OS X) and our IT department is reluctant to support them so we're doing most of the support for them ourselves. So for now we fly by the seat of our pants a bit.

    27. Re:HIPAA Constraints? by kramulous · · Score: 1

      In case you need help convincing the hierarchy and you need a little ammunition to get a decent, scalable, centralised solution, you will find allies in:
      Engineering - find out those that teach and apply for grants doing any kind of FEA work, the robot people,
      Physics: The medical imagers, users of geant4 and beam, biomechanical,
      Comp Sci: talk to anyone related to the document searching/indexing areas, machine learning, etc
      Chemistry: Search you local paper repository for those that have someone from your maths department/school as an author with a chem bod - all sorts of partnerships there
      Mathematics: find the algebra buys.
      Health: You will have people trawling data on world health that use a small machine and SAS, SPSS. They will need your help

      just to name a few.

      Good luck man.

      --
      .
    28. Re:HIPAA Constraints? by dogmatixpsych · · Score: 1

      Thanks for your comments and advice.

    29. Re:HIPAA Constraints? by rwa2 · · Score: 2, Interesting

      Yeah, then it sounds like you're pretty much doing the best you can under the circumstances... I was just trying to think out of the box a bit and turn your filesystem compatibility problem into a file server compatibility problem, since cross-platform compatibility is a much bigger deal in the latter scenario.

      One last consideration you might want to try benchmarking is storing your data in an image file, like a zip or tgz or more likely a dmg archive... that way you could probably do transparent compression as well, which might work well on your datasets. If your disks have limited performance from USB or whatever and you have beefy CPUs, that may even increase your overall throughput when making copies to/from fast local storage. Plus some of the archival formats may help preserve some filesystem features / metadata that may otherwise get clobbered by copying through a compatible intermediary FS.

    30. Re:HIPAA Constraints? by Qzukk · · Score: 1

      HIPAA seems to be part "let's make an attempt - it doesnt matter if it's a good one" and part "this should appease the non-technical public, who wont know any better"

      Ding ding ding!

      Can't keep count of the number of people who send me an email declaring that I have a secure message waiting for me! Just follow this un-password-protected link to our web site and read it without any further authentication other than having read the link from my email. Oh, but the site uses SSL, so it's secure !

      --
      If I have been able to see further than others, it is because I bought a pair of binoculars.
    31. Re:HIPAA Constraints? by dogmatixpsych · · Score: 1

      Hmm, those are good suggestions. We'll (I'll) have to give them a shot. Most of our smaller image files are gzipped but these large files have not been (largely because not only would they take a long time to gzip with probably only minimal benefits but also we're just starting to create them and haven't had to "store" them much). Anyway, you make some great suggestions.

    32. Re:HIPAA Constraints? by mlts · · Score: 1

      TrueCrypt is probably the only block encryption system that is compatible with Mac, Linux, and Windows. I don't think it would work in the OP's case, but if I were moving sensitive data between these three platforms, I'd be using it, and a keyfile on a smart card. This way, if the drive and smart card were snatched, an attacker would have 3 guesses to guess the 20+ character on the smart card before it erased itself, then they would still have to brute force the main TC key to get access to the volume.

  7. Don't know if broke by tepples · · Score: 2, Informative

    More like "I don't know if it's broke. I could be doing something so wrong it could end up on The Daily WTF. If what I'm doing is broke, could you help me fix it?"

    1. Re:Don't know if broke by Anonymous Coward · · Score: 0

      More like "I don't know if it's broke. I could be doing something so wrong it could end up on The Daily WTF. If what I'm doing is broke, could you help me fix it?"

      I'm afraid that's considerably less likely to catch on as a slogan.

  8. FAT32 -NOT by Anonymous Coward · · Score: 0

    FAT32 has a file limit size of 4GB.

    He needs 20 times that!

    1. Re:FAT32 -NOT by raynet · · Score: 1

      Better just use RAR and compress them with recovery record, multipart and password.

      --
      - Raynet --> .
    2. Re:FAT32 -NOT by X0563511 · · Score: 1

      No problem, that won't take any time at all...

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    3. Re:FAT32 -NOT by Wolfraider · · Score: 0

      Are we saying FAT32 is not so FAT after all?

    4. Re:FAT32 -NOT by Anonymous Coward · · Score: 0

      FAT + DriveSpace!

    5. Re:FAT32 -NOT by cynyr · · Score: 1

      then skip the compress(and rar) and just use a small bash script and dd. no slower than writing it over USB anyways.... 80GB over USB2 must be painful.

      --
      All of the above was encrypted with a Quad ROT-13 method. Unauthorized decryption is in violation of the DMCA.
    6. Re:FAT32 -NOT by raynet · · Score: 1

      I prefer Stacker and their memory compressor for Windows 3.11 was awesome too. Still this wouldn't solve the maximum filesize issue that multipart volumes with RAR does solve.

      --
      - Raynet --> .
    7. Re:FAT32 -NOT by X0563511 · · Score: 1

      Yea, it certainly would.

      I found an eSATA bracket that plugges into a normal SATA port on the motherboard. It fit just fine on the back of my server's supermicro 1U chassis.

      My external drive has an eSATA as well. I've got tested sustained data rates over 100mb/s - definitely superior to USB in that respect.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
  9. FAT32 is a fucking horrible idea in his case. by Anonymous Coward · · Score: 3, Informative

    How the fuck is he supposed to store 80 GB files on a filesystem that maxes out at 4 GB?

    1. Re:FAT32 is a fucking horrible idea in his case. by Tirs · · Score: 3, Funny

      Sorry man, but your use of the "f" word is totally inadequate in this conversation. Let me correct you:

      How the fsck is he supposed to store 80 GB files on a filesystem that maxes out at 4 GB?

      Much more in-context, eh?

      --
      Strength, balance, courage and reason. If you know what's this about, contact me!
    2. Re:FAT32 is a fucking horrible idea in his case. by Anonymous Coward · · Score: 1, Funny

      Go fsck yourself, diskhead.

    3. Re:FAT32 is a fucking horrible idea in his case. by mkiwi · · Score: 1

      How the fuck is he supposed to store 80 GB files on a filesystem that maxes out at 4 GB?

      It's called liposuction.

  10. NTFS by Trevelyan · · Score: 3, Interesting

    NTFS or any other FUSE (MacFUSE) file system. However in a heterogeneous environment NTFS has the bonus of native Windows support.

    There is NTFS-3G for Linux and Mac OS X

    There is also an EXT2 Fuse FS (for Mac OS), and probably many other options.

    Having said that, I have never had a problem with Linux's HFS+ write support.

  11. ext2 works. ntfs works. by Snover · · Score: 1

    Mac OS and Linux both have support for NTFS through NTFS-3G. Mac OS has support for ext2 through fuse-ext2.

    --

    [insert witty comment here]
    1. Re:ext2 works. ntfs works. by EXrider · · Score: 1

      Write performance through FUSE on Mac OS X is pretty disappointing, several orders of magnitude slower than direct filesystem access in my experience. Transferring an 80GB file to a FUSE mounted filesystem would be painful.

      --
      grep -iw skynet /etc/services
    2. Re:ext2 works. ntfs works. by MachineShedFred · · Score: 3, Informative

      If it's Mac OS X 10.6.x, you don't even n eed NTFS-3G, as the native NTFS driver has read / write capability. You just need to change the /etc/fstab entry for the volume to rw, and remount.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    3. Re:ext2 works. ntfs works. by Anonymous Coward · · Score: 5, Informative

      This is dangerous advice. There are numerous reports of instability and NTFS volume corruption when forcing 10.6 to mount NTFS volumes R/W. Apple seems to have turned NTFS write off by default for a good reason, it's not done yet.

    4. Re:ext2 works. ntfs works. by mehemiah · · Score: 1

      they both have support for hfs+

    5. Re:ext2 works. ntfs works. by Snover · · Score: 1

      But, HFS+ in Linux is read-only. Unless you force it, in which case journaling is disabled and Mac OS freaks out and does a full filesystem check the next time you connect it.

      --

      [insert witty comment here]
    6. Re:ext2 works. ntfs works. by Anonymous Coward · · Score: 0

      Because Apple doesn't have a license to use NTFS? That's most often what I've read. I've used NTFS r/w in OSX 10.6.3 via the ntfsmounter app and I haven't come across any problems yet. Mainly I'm copying system images of a few gigs to an external firewire disk for backup which is formatted NTFS.

    7. Re:ext2 works. ntfs works. by Anonymous Coward · · Score: 0

      I can second this. I have been transferring files over from an old system to a new Mac, and I found that it was much faster to read the external disk in NetBSD running under VMware Fusion and then write to a nfs mounted shared Mac folder, than using any of ntfs, ntfs-3g or fuse-ext3 directly under Mac OSX to do the copying.

      I just don't think that a certain person at Apple wants users to use anything other than HFS+ ;)

    8. Re:ext2 works. ntfs works. by Anonymous Coward · · Score: 1, Interesting

      "Works for me" != "works for everyone". Especially if your usage patterns don't happen to trigger the bugs...

      Also, you do not ned a license to use NTFS. ntfs-3g is a free open source implementation based on reverse engineering. It's been around for a long time and Microsoft hasn't shut it down. As far as I know, Apple's NTFS driver is also based on reverse engineering. It's just notoriously hard to implement writing to NTFS; it's a complicated file system.

  12. HFS+ unjournaled is best; MacFUSE also works by vossman77 · · Score: 1

    I have a similar scenario and I think HFS+ unjournaled is best for your scenario. FAT32 is even worse. You are fortunate not to have to support windows. Ideally I would use NFS and file sharing instead of external disks. But shipping a disk is always better than transferring large amounts of data over the net.

    Another option is to install MacFUSE and then mount other file systems. This is what I do when NTFS is required. For my Linux system I love ext4, if you need an older file system use XFS, ext3 is stable but really slow for large data files.

    1. Re:HFS+ unjournaled is best; MacFUSE also works by toQDuj · · Score: 1

      Another vote for NFS here. It is truly shameful that after 60+ years of computing, there is still no interoperable file system around. Network storage is your solution there, although transfer speeds are seriously limiting then.

      --
      Every experiment which ends in a big bang is a good experiment.
  13. NTFS by dirtyhippie · · Score: 1

    It sucks, but NTFS might just be the best option. OSX and linux both have had stable enough support for years. The main plusses over FAT32 are journaling and support for files > 4GB. Using UFS is dangerous (or at least has been until very recently) because there are so many different variants of it (solaris, BSD, osx, etc.) that linux support is notoriously troublesome. An extra plus of NTFS is you can use it easily on windows machines as well.

  14. Reiser? by Wowsers · · Score: 4, Funny

    I would have recommended ReiserFS, but the data might get buried somewhere and the system would not remember where it was....

    --
    Take Nobody's Word For It.
    1. Re:Reiser? by Anonymous Coward · · Score: 4, Funny

      That's pure pure FUD. ReiserFS can recover anything, even something it allegedly never stored. "Oakland homicide detective Lt. Ersie Joyner recalled that Reiser led them directly to the exact site, without any hesitation or confusion."

    2. Re:Reiser? by Haxzaw · · Score: 0

      Woooosh!

    3. Re:Reiser? by quickOnTheUptake · · Score: 1

      whoosh

      --
      Mod points: Guaranteed to remove your sense of humor.
      Side effects may include gullibility and temporary retardation
    4. Re:Reiser? by Anonymous Coward · · Score: 0

      so funny I forgot to laugh

    5. Re:Reiser? by MarkRose · · Score: 1

      Didn't you read the summary? He's transporting the data in his car. Don't you remember what Reiser does to car seats? It's get them all bloody and ejects them! Didn't you think this through? What if he is driving when the seat gets ejected??

      --
      Be relentless!
  15. No Filesystem by Rantastic · · Score: 5, Informative

    If you are only moving files from one system to another, and do not need to edit them on the portable drives, skip the filesystem and just use tar. Tar will happily write to and read from raw block devices... In fact, that is exactly what it was designed to do. A side benefit of this approach is that you won't lose any drive capacity to filesystem overhead.

    --
    Ask Slashdot: Where bad ideas meet poor googling skills.
    1. Re:No Filesystem by Anonymous Coward · · Score: 0

      He might not need to edit them in place, but he almost certainly needs to be able to directly read the files from the portable disk, without transferring them again. And if it's not a definite need, it would be a convenience well worth keeping.

    2. Re:No Filesystem by Anonymous Coward · · Score: 0

      Genius! Simple, as fast as can be and utilizes every last bit if needed. I overlooked this approach when thinking of what filesystems would be best. Never crossed my mind that you could ditch the filesystem entirely.

    3. Re:No Filesystem by Anonymous Coward · · Score: 0

      Why do you think he needs to directly read the file from the portable drive? He says specifically in the question that they're transferring the images from one systems in one labratory to systems in another labratory.

      He's not asking for a storage system, or an archive system, or the best way to share documents. He wants to read raw bits off of permanent storage in one laboratory into a temporary repository, write those bits from into permanent storage in another laboratory, and then destroy the temporary storage.

      With that said, tar is a bad solution because it doesn't include any type of CRC or encryption. But it's a good idea, and certainly a million times better than a file system of some type.

    4. Re:No Filesystem by Rantastic · · Score: 2, Informative

      With that said, tar is a bad solution because it doesn't include any type of CRC or encryption. But it's a good idea, and certainly a million times better than a file system of some type.

      True, but simply hashing the file at both ends solves that. Both linux and mac support shasum.

      --
      Ask Slashdot: Where bad ideas meet poor googling skills.
    5. Re:No Filesystem by Rantastic · · Score: 2, Informative

      As to encryption, you just encrypt the file before you tar it. In fact, with gpg you get both encryption and integrity checking.

      Gnupg is available in Mac Ports and comes with just about every linux distro.

      --
      Ask Slashdot: Where bad ideas meet poor googling skills.
    6. Re:No Filesystem by aix+tom · · Score: 1

      Because he said he has to "readily read and write to them".

      I suspect the way a laboratory works, he wants to plug the disc in and then look / work on the data without first having to copy it to the local drive all the time in a lot of cases.

    7. Re:No Filesystem by toQDuj · · Score: 1

      I had no idea you could do that with tar. I will be looking into that for my next mass-transfer problem!

      --
      Every experiment which ends in a big bang is a good experiment.
    8. Re:No Filesystem by ignavus · · Score: 1

      A side benefit of this approach is that you won't lose any drive capacity to filesystem overhead.

      Well, TAR becomes your filesystem, as it includes basic filesystem information about file names and locations (and paths and permissions and...).

      Of course, you could just write one file to a disk, starting at the first sector. Then you really would have no filesystem overhead.

      --
      I am anarch of all I survey.
    9. Re:No Filesystem by rdnetto · · Score: 1

      If you had told me at the beginning of this post that the best solution was to simply role your own filesystem, I would have laughed.

      --
      Most human behaviour can be explained in terms of identity.
    10. Re:No Filesystem by dlgeek · · Score: 1

      Well, keep in mind that using tar can cause all kinds of sticky problems when transferring masses. I'd suggest some kind of catapult or trebuchet instead.

  16. ext3 or ext4 by ecloud · · Score: 1

    Here's a commercial $39.95 implementation of ext2/3/4 for MacOS. No idea if it's actually any good. I'd really like to hear if someone here has tried it, because I might like to use it for a shared /home between Linux and MacOS if it would work. I tried hfs+ (or it was it just hfs?) without journaling, and the dang thing needs to be fsck'd nearly every time I booted the alternate OS, which wastes a lot of time. Particularly, when shutting down Linux it's unmounted cleanly (such that Linux is happy if I just boot back into Linux again), but MacOS is still not happy and does the fsck in the background for 15 minutes or so before I can access it again. Sometimes it fails too, and has to be done manually from Disk Utility. Quite annoying.

    1. Re:ext3 or ext4 by mlts · · Score: 1

      I have heard good things about that, and I think the last college I went to had that in use so the Windows boxes would be able to grok HFS+.

      I say slap that on the Mac and call it done. Since it is a commercial product, if there any glitches, you can blame it on that product, so the $40 pays for some CYA.

  17. Rubbish by Improv · · Score: 4, Informative

    You're storing it in the wrong format - there are all sorts of tools to convert to Analyse or DICOM format, which give you a managable frame-by-frame set of images rather than one huge one. Most tools to manipulate MRI data expect DICOM or Analyse anyhow (BrainVoyager, NISTools, etc).

    If you really want to keep it all safe, use tarfiles to hold structured data, although if you do that you've made it big again.

    Removable media are a daft long-term storage - use ad-hoc removable media solutions (or more ideally, scp) to move the data.

    --
    For every problem, there is at least one solution that is simple, neat, and wrong.
    1. Re:Rubbish by dogmatixpsych · · Score: 1

      Our large files are not DICOM or ANALYZE or NIfTI files. All of those are small; if that's all we dealt with we would not have this issue. Our large files are fiber tracking files that are a particular format for a particular visualization and analyzation program. In any case, the only files going on these drives are easily re-creatable should the drives explode or something like that (it would take time but we only want to put end-result files and keep all the other files that are used to create them on our managed storage).

    2. Re:Rubbish by sn00ker · · Score: 1

      Oi. You've been around long enough to know the rules. Knowledgeable, informed posts are contrary to the T&C. Go back to your porch. Sheesh. It's geezers like you that give whippersnappers like me a bad name. I'll be getting off your lawn now.

      --
      "God, root, what is difference?" - Pitr, userfriendly
    3. Re:Rubbish by Improv · · Score: 1

      Oh - maybe the original story wasn't clear enough about what these were - once one moves into the analysis stage, it's quite possible to make larger intermediary files, as you have. The best advice I could give would be to use VFAT in that case, unfortunately.

      --
      For every problem, there is at least one solution that is simple, neat, and wrong.
    4. Re:Rubbish by dogmatixpsych · · Score: 1

      Yeah, I was trying to not write too much in the original post because most Slashdotters don't read it anyway. :) Yes, these large files are intermediary and end-stage analyses.

    5. Re:Rubbish by __aasqbs9791 · · Score: 1

      How long does it take to create the fiber tracking files? Though it probably won't work for you, if the file size is very different between the raw images and the fiber tracking files you might be better off to transfer the raw images and create the fiber tracking files at the destination, rather than creating and then transporting them. I used to work with large numbers of smaller files (millions of TIFF images transfered to different clients in different naming conventions and formats) a couple of year ago and we had to transfer the files via drives as well. 4+ TB just doesn't transfer very well, even over business class internet, so I have some idea what kind of pain you are under (we didn't have to deal with HIPAA though, since these were public county records).

    6. Re:Rubbish by dogmatixpsych · · Score: 1

      That's actually what we will probably do since you can take the raw files and (at least on our Macs with i7 processors) recreate the fiber tracking files in about 20 to 60 minutes. Because of this, we might not even hang on to these large files because they can be recreated easily. However, it's always nice to have the "original" file with research should anyone want to see it (and not just a recreated file), which is why we probably will hang on to them (plus, it does take time to run the tracking).

      Your comment is well made and insightful.

    7. Re:Rubbish by __aasqbs9791 · · Score: 1

      Before I started that job, I thought I knew what big was. Millions of something was a concept I thought simple enough. Then I actually had to deal with them, and transfer them around, verify integrity, and rename them (in some cases) and I now know I didn't know what a TB of small files looked like. Not really. And since we had 4 customers that all wanted something different (some wanted separate files for each page, some wanted them combined into a single file per document, some wanted 4 digits for the year, some wanted 2 digits, unless it was below a certain year, etc) and I had to make that all happen. And now I look in awe at Petabyte sized file references because I know I can't actually grasp how to handle that kind of data.

  18. Qualifications by ecloud · · Score: 1

    I'd say the "best format" needs journaling absolutely, and preferably also extended attributes which work consistently between the two OS's, hardlinks and symlinks working consistently, long filenames, case sensitive, separate metadata for creation time and modification time, suitability to be used on a USB flash drive as well as a hard disk, and ability to mount it in Windows too. Haven't found any such mythical beast yet. If somebody would just finish the journaling support for Linux HFS+....

  19. Who cares? by rjstanford · · Score: 1

    No, seriously, who cares? This is a process designed to save files that are then transferred through SneakerNet. While moderately large, at 80gb, they're not huge by modern standards. If you have a current solution that works, stick with it.

    If, however, there are other constraints that are affecting you - transfer speed, decades-long retention on local media, security, etc, then by all means let us know. Until then, to use the obligatory car analogy, its as if you've said:

    Due to the distance between my house and work, I currently use an automobile to go between the two locations and to perform various other services. Currently I use a Honda Accord. What would you suggest?

    --
    You're special forces then? That's great! I just love your olympics!
  20. NFS over SSH by HockeyPuck · · Score: 2, Interesting

    Just tunnel NFS over SSH. I can't imagine how secure it would be to sneakernet any files around the office. If you need to encrypt the data at rest then either encrypt on the client or leverage an encrypted filesystem of a Decru type appliance.

  21. Encrypt by Anonymous Coward · · Score: 0

    If you are carrying these things around, I would think setting them up with TrueCrypt (works on OSX and linux) or other full disk encryption program would be a Good Idea (TM). It would be bad to have sensitive data get left somewhere. This way you would just be out the cost of hardware (assuming the data is backed-up). I would guess that HIPAA would want you to do what you can to protect that from happening.

    Best of all, it has a proven track record: http://www.theregister.co.uk/2010/06/28/brazil_banker_crypto_lock_out/

  22. 4GB per file limit by Ilgaz · · Score: 4, Insightful

    OS X UFS has a very unfortunate limit as it doesn't support files over 4 GB. Or, there was no chance, I would format everything (especially USB) as UFS.

    Lack of commercial quality disk tools like Disk Warrior if a true catastrophe happens is a problem too. Of course, fsck can do good things but after a true catastrophic filesystem issue, diskwarrior is a must. That was one of the things Professional Mac community had hard time explaining ZFS community.

    As Apple was truly wise to completely document it down to a point you can even write a full feature defragmenter (iDefrag), HFS+ without journaling seems to be the best option. I am in video business and I have seen it deal with files way beyond 80GB without any issues. In fact, lots of OS X users who images their drives see it everyday too.

    I don't know why journaling is not implemented, it is open and documented too. If a bit hassle happens, it sure deserves it since he deals with external drives which are just fit to journaling purposes.

    1. Re:4GB per file limit by Larryish · · Score: 1, Funny

      iThis and iThat, blah blah blah

      How about iJustpukedalittlebitinmymouth?

    2. Re:4GB per file limit by Anonymous Coward · · Score: 0

      I have actually managed to break the HFS+ file system. I accidentally streamed three backup servers to the same RAID simultaneously. The three files created all surpassed 1TB in size and soon enough the fragmentation got so bad any writes what-so-ever to the drive took about 10 minutes. That's how robust HFS+ is: you can screw it up, but you have to be a complete bonehead with outrageous sized files to do it.

  23. FAT32 is a nightmare waiting to happen by Ilgaz · · Score: 1

    Most of files they produce involves an actual patient, sometimes in critical condition stay in something like a grave for hour sometimes.

    If one of issues with filesystem, that archaic junk which should have never been released happens, it will be nightmare to restore the data while it is easy on HFS+ Journaled or even NTFS.

    I own a Symbian phone and trust me on that, if there was a $50 utility just to get rid of FAT32(!) junk risking my data on memory card, I would happily buy it.

  24. X-Ray and iPod? by Ilgaz · · Score: 1

    I heard it is almost a standard procedure to use iPods in X-Ray community for the x-ray format images and that is why there are several OS X Utilities supporting it.

    I guess the first reason was the gigantic (for that time) storage size of the iPod and you can also use it for music.

  25. Network? by guruevi · · Score: 4, Informative

    Really, you need a gigabit network and transfer files over it using AFP and/or NFS and/or SMB. First of all HIPAA requires you to encrypt your hard drives which most researchers won't do (it's too difficult). Then you also got the problem what happens if the researchers (or somebody else) leaves with the data.

    Solaris and by extension Nexenta have really good solutions for this. You can DIY a 40TB RAIDZ2 system for well under $18,000. If you use desktop SATA drives (which I wouldn't recommend but ZFS keeps it safe) for your data you can press that cost to $10 or $12k.

    I work in the same environment as you (neuroimaging, large datasets), feel free to contact me privately for more info.

    --
    Custom electronics and digital signage for your business: www.evcircuits.com
    1. Re:Network? by dogmatixpsych · · Score: 1

      We've talked with our privacy office and our files do not have to be encrypted because they are deidentified from the start. The privacy office of course still prefers that they are encrypted (which we will do) but in our case with our scans (the ones going on these drives are impossible to identify someone with and do not strictly count as personal health information). Everything is kept secure and we will encrypt just to be super safe.

    2. Re:Network? by Anonymous Coward · · Score: 0

      Backblaze StoragePod for the win! Well, you would have to fiddle with stuff and and the design isn't ideal with the port multipliers, but it's good enough for most folks. You could concentrate it all onto an Areca 16 port RAID or JBOD card (you end up using 9 ports), since besides the ZFS part there is nothing redundant anyways.

    3. Re:Network? by markdavis · · Score: 1

      >First of all HIPAA requires you to encrypt your hard drives

      That is 100% *WRONG*. There is absolutely NO such requirement (and yes, I know, because I am a HIPAA Security Officer). You are, however required to analyze/discuss, document, and follow certain procedures.

    4. Re:Network? by Anonymous Coward · · Score: 0

      Sorry, wrong. HIPAA _does not_ require hard drive encryption as a general rule. Encryption is required only if the data will be handled by entities who are either not authorized to view patient data (e.g transferring it over the unencrypted internet via ISP's or transfer by courier) or if the data will be transported where it is likely to fall into the hands of un-authorized persons (e.g on removable media, external drives or laptops taken outside the facility).

      As long as sneakernet is used internally to the site working with the data and no un-authorized persons have access to the data, encryption is not required by HIPAA.

    5. Re:Network? by Anonymous Coward · · Score: 0

      and how do you dispose of PCs?

      I'll take that crestfallen look on your face as acceptance that you're non-compliant. Scurry away, little mouse, and go tell your bosses the bad news.

      Using encrypted hard disks means that so long as the disposal company doesn't have the key (and why would they?), the disks are unimportant.

      Most operating systems you'd actually use on the desktop (including desktop Linux distros) offer encrypted disk as an option. It should be installed this way by IT, and the users told the password.

      In 2010 practically everyone should be doing this, not just for HIPAA but because it just makes sense.

  26. NTFS is undocumented and read only on OS X by Ilgaz · · Score: 1

    As Apple didn't want to bother with "OS X deleted my NTFS drive" people, they support NTFS as read only by default. "NTFS-3G" and other utilities can of course read/write but it doesn't change the true reason for NTFS being unreliable to support: It is _not_ documented.

    HFS+ on the other hand is completely documented. Apple wins on this case because of openness and the fact that, their true discipline in making things backwards/forwards compatible with complete documentation.

    Nobody had to/has to reverse engineer anything, the source code of HFS+ is right at opensource.apple.com

    1. Re:NTFS is undocumented and read only on OS X by dgatwood · · Score: 1

      UDF seems like it would be a good choice to consider. It's natively writable in Mac OS X v10.5 and later, Linux 2.6 and later, and Windows Vista and later.

      --

      Check out my sci-fi/humor trilogy at PatriotsBooks.

    2. Re:NTFS is undocumented and read only on OS X by marquise2000 · · Score: 1

      I'm using UDF for my LinuxOSX setting and it's awesome.

      --

      Marquise

      -- I need a new sig.

    3. Re:NTFS is undocumented and read only on OS X by WhiteDragon · · Score: 1

      I agree with the parent. UDF supports large files (up to 16 EB) and as the parent said, is writable on most modern OSs

      --
      Did you mount a military-grade, variable-focus MASER on an unlicensed artificial intelligence?
  27. Best format is... by Picardo85 · · Score: 0

    I would say - 3,5" if it's a desktop or 2,5" if it's a notebook

  28. Samba that is by wasabioss · · Score: 1

    You should consider creating a samba share. If you don't have a server a NAS appliance or a el heapo old PC would handle the job very well.

  29. Network! by Anubis350 · · Score: 1

    RAID array and NFS, a Lustre, etc depending on need, but a network share! ...and if you need more encryption and even admins cant have access to data, have your users store true-crypt drives on the network. Sneakernet is, in the end, far more insecure!

    --
    "goodbye and hello, as always" ~Prince Corwin, from Zelazny's Amber series
  30. UDF by marquise2000 · · Score: 2, Informative

    I'm using a USB Disk formatted under linux with UDF (yep, it's not limited to DVDs, there is a profile for hard disks). It can be used without problems under OSX (even Snow Leopard)
     

    --

    Marquise

    -- I need a new sig.

  31. XFS? by Anonymous Coward · · Score: 0

    XFS handles large files VERY well, but I'm not sure of OSX support. Darwin is just a modified BSD kernel so it COULD have support. Question is whether the iCzar decided to build it in . . .

  32. Use tar, that's what it is for by Anonymous Coward · · Score: 0

    tar --posix --noxattrs --mode 644 cvf /dev/... honking-big-file.img

  33. NTFS by mstrebe · · Score: 1

    Handles large files well, doesn't respect UNIX permissions (which are problematic for removable drives) and is freely installable on Linux and Mac using NTFS3G and FUSE or MacFUSE. A lo

    --
    aka Matthew at SlashNOT/!
  34. the question of our age by commodoresloat · · Score: 4, Insightful

    who will wooosh the woooshers?

    1. Re:the question of our age by Anonymous Coward · · Score: 0

      all my mod points are belong to you.

  35. NAS device by linebackn · · Score: 2, Insightful

    A simple NAS enclosure or NAS device might be what you are looking for. You can get a single drive NAS enclosure, and add a drive, that you can carry around just like a regular portable drive. You can move it between networks and use any connection method the NAS device happens to implement (SMB, FTP, NFS, etc). Some even let you optionally connect it directly via USB or eSATA to access the file system directly, and some may have encryption or other security features as well.

    Of course, check to make sure you have permission and that connecting things to your network does not violate any policies. If connecting a network device directly to the your network is not permitted then perhaps you can add a second, dedicated, network card to the computers.

    1. Re:NAS device by EEPROMS · · Score: 1

      + 1, why stuff around with a filesystems when you can get a small compact single or dual disk NAS that is specifically designed for storing large chunks of data and wont unlike USB based externals disk randomly turn your data into junk. I have a small dual disk NAS with two 2.5" hardisks setup in raid 1 (thus if one disk dies I have another with a copy) that I can ssh/ftp from any PC (also my mobile phone if you have WiFi) on the network. It does have a USB port as well as a LAN socket so I can still use it as an external USB disk if need be.

    2. Re:NAS device by EEPROMS · · Score: 1

      forgot to mention in USB mode it shows up as a Fat32 device (not the actual file system in use though, in my case the NAS uses XFS) the fat32 transfer limits still apply but feeds the data stored out in chunks to get around this.

  36. do not use a filesystem by harlows_monkeys · · Score: 1

    Treat the disk as if it were a tape, and use the GNU version of cpio.

    You can install GNU cpio via macports on your Macs, and people with Linux should find it either already installed or available in their distribution's package system.

    You need to use the GNU cpio instead of the BSD cpio that ships with OS X because there are incompatibilities between the two, and I was unable to find a set of settings that would make them compatible. (There are settings that should, but they did not work, so there's a bug in one or the other--or both. However, after a few minutes experimenting it occurred to me to just grab GNU cpio from macports and use it on both ends, so I never pursued tracking down the incompatibility)

    Note that you can also get GNU cpio for Windows.

  37. filesystem is largely irrelevant by drfireman · · Score: 1

    I do my fair share of transferring large neuroimaging datasets around from time to time, although I don't do it regularly. If you want to use hard drives that aren't connected to anything in transit, then I have to agree with whoever suggested doing it without a filesystem. I've always found that to be the easiest way to get around filesystem (and sometimes operating system) idiosyncrasies, whether you're writing to a DVD or a hard drive or whatever. If you can (de)serialize your data easily (using tar), then a filesystem does almost nothing useful for you, it just gets in your way.

    I certainly won't agree with the respondent who suggested that you change your workflow (and perhaps your choice of tools) to use smaller files. If those are your files, those are your files. Perhaps I'd do it differently, but that's my problem, not yours.

  38. not a question of file system as such - use a NAS! by hherb · · Score: 1

    I have a similar problem with backups in my paper less medical practice - I always need a working system off-site for emergency replacement, and here in rural Australia doing it via Internet is impossible due to lack of networking infrastructure and ridiculous bandwidth costs
    I use a QNAP NAS (TS659). They also come as tiny handy cubes with 2.5" disks instead of the 3.5"
    That makes the question of the file system irrelevant, since it communicates with just about any operating system through standard protocols (eg NFS, AFP, SMB). Internally they use ext3 or ext4 file system depending on user preferences. Whenever I want to transfer files, I simply take one disk drawer out and simply shove it into the QNAP at the second location, where it will immediately start to replicate it. I use a spare disk for the original system, where the RAID1 configuration will immediately rebuild it, to be ready when I come again for the disk next day.

    Horst

  39. Re:UDF IS ACTUALLY A SOLUTION by marquise2000 · · Score: 5, Informative

    Ok everybody's occupied with surreal suggestions, but anyway:
    *UDF* is quite awesome as a on disk format for LinuxOSX data exchange, because it has a file size limit around 128TB, supports all the posix permissions, hard and soft links and whatnots. There is a nice whitepaper summing it all up:
    http://www.13thmonkey.org/documentation/UDF/UDF_whitepaper.pdf

    If you want to use UDF on a hard disk, prepare it under linux:
    1) Install uddftools
    2) wipe the first few blocks of the hard disk, i.e. dd if=/dev/zero of=/dev/sdb bs=1k count=100
    3) create the file system : mkudffs --media-type=hd --utf8 /dev/sdb (that's right, UDF takes the whole disk, now partitions)

    If you plug this into OSX, the drive will show up as "LinuxUDF". I am using this setup for years to move data between linux and OSX machines.

    --

    Marquise

    -- I need a new sig.

  40. NFS (no wait, I DID read it) by Sloppy · · Score: 1

    The best cross-platform (Linux+MacOS) filesystem is NFS, wh-- stop hitting me, I DID read the whole question. Ok? So, as I was about to say, use NFS. When the techno-ignorant HIPAA people watch what you're doing, just send 80gig of /dev/random (bonus: it looks encrypted, the HIPAA guys will love that) to the removable drive, and when you're coping off that drive, send to /dev/null. Meanwhile, as the drive's contents are going to your lame software-emulated null device, also be reading the file off the ethernet via NFS. Problem solved.

    But in all seriousness, the guy who said "tar" was right. I wouldn't have thought of that. Brilliant.

    --
    As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
  41. Re:UDF IS ACTUALLY A SOLUTION by fnj · · Score: 1

    Give the man a cigar. I was struggling through all the other suggestions, every single one of them involving unacceptably horrible tradeoffs, and finally get to this post, the only idea that is not just mind numbingly brain dead. I don't even use OSX any more (finally cured that brain disease), and I'm gonna check this out.

  42. NTFS by led_belly · · Score: 0

    journalling is there to help provide a safe guard against data loss and corruption. using HFS+ without the journal can be risky - I have experienced problems personally from this - any freeze up, reboot without proper unmounting of the drive, etc and you are in trouble. While I haven't tried NTFS is seems to be the most compatible between OS's, meets your requirements for file size of 80 GB's:

    https://secure.wikimedia.org/wikipedia/en/wiki/Comparison_of_file_systems

    A FREE driver is required:

    https://secure.wikimedia.org/wikipedia/en/wiki/NTFS-3G (this driver seems well tested and stable)

  43. Compression? by randallman · · Score: 1

    Just curious. Is the 80GB after applying lossless compression to the image set? If not, there's no good reason to store it uncompressed.

    As for your question, I agree with those that say to skip the filesystem. Just use tar and a block device.

    -Randall

  44. Depends on what OSX supports by GoochOwnsYou · · Score: 1

    Bit of a pickle, last time I checked OSX doesn't support ext4 while Linux doesn't support ZFS (except with FUSE which I dont reccomend in a produciton environment) which would be my 2 options for performance. I know UFS is in both so that could work but I am curious if OSX can support XFS.

    If you could find a driver for ext4 on OSX I would reccomend that: its fast & proven to be quite stable, next best thing would be XFS.

    --
    This sig has been distributed under the Creative Commons license.
  45. What is wrong with using a supported NTFS on OS X? by ssj152 · · Score: 1

    I see several mentions of using NTFS on OS X but no mention of using a commercial SUPPORTED version that is commonly available - for instance Paragon NTFS 8. I have used v7 and upgraded to v8 and had little to no issues with it. The only problem I've had was with a corrupted volume.

    Is this a religious or moral issue - using only FOSS - or am I unaware of significant issues with this product? It would seem to me that a business would be willing to pay a small amount for support on a product.

    I have no interest in the product other than being a user of it.

    --
    Be Obscure Clearly
    There are visual errors in time as well as in space.
  46. YES FAT32 is a nightmare waiting to happen by Anonymous Coward · · Score: 0

    We were forced by a customer to use it for a video storage project. Things just
    get lost, never to be found again. ext3 w/ data=journal has been bulletproof
    with pretty good retention of the last few minutes. So far.

  47. Mac Fuse- Google project by acomj · · Score: 1

    Something to look into.. Allows other file systems to be used with a mac.

    http://code.google.com/p/macfuse/

  48. Re:UDF IS ACTUALLY A SOLUTION by Anonymous Coward · · Score: 0

    This is a good solution.

    I format my USB flash cards with UDF to exchange large files between Mac OS X and Linux (ya' just can't beat sneakernet!).

    I've not (yet) been able to get this to work with Windows though. Even though "modern" versions of Windows supposedly support UDF, it seems to be for optical media only.

  49. Re:UDF IS ACTUALLY A SOLUTION by mlts · · Score: 2, Insightful

    That is an excellent solution, and arguably the best to the OP's problem printed. UDF works on Windows, OS X, Linux. Even AIX is happy with it and can write to it. So an external drive with this on it should definitely solve the problem.

  50. biz opportunity by zogger · · Score: 1

    I always like to think of the profit angle when there is a problem, the silver lining hiding in plain sight with any problem or hassle. Your situation sounds like a great way to go into business for yourself. Walk away with a few other key players, then contract the work for more money-then do it, use the machines you really need, and as long as you are forced into being your own IT department-so what? You are *anyway* according to what you say so you might as well get paid for it. Get mo' money for more responsibility. Not sure how this works with grants, but sounds like the work really needs to be done, and medical stuff....seems like anything medical related today is a license to print money somewhere just under top investment banks.

    File systems...seems like all the heavy hitters here are always going on and on about ZFS, best thing since burritos to go in a bag, etc, so is that an option for your macs and the recipients?

    As to file sizes..I'm a dumbass on this but can't they be chunked up, like torrent files are chunked up into much smaller sizes? Torrent them to..wherever they need to go to?

    1. Re:biz opportunity by dogmatixpsych · · Score: 1

      I'll have to take the business plan under consideration. As far as breaking up the files - yes we could but the issue is with multiple IT departments and storage on either end of the transfer. Plus, I didn't include this in my initial question post but much of the time the transferring isn't permanent meaning that it is more of "showing" the files than giving the files. In other words, our collaborators want to see the files (images) but not hang on to them, which is why it seems to make more sense just to use external HDDs. Plus, it only takes us 5 minutes to walk over there so it's no big deal to run back and forth should we need to.

  51. A Plethora of Possibilities... by MrWin2kMan · · Score: 1

    FAT32, NTFS, perhaps ZFS...and following up to the HIPAA remars above, I would use TrueCrypt as well.

    --
    Nothing to see here but us trolls...move along...
  52. five minute walk by zogger · · Score: 1

    OK, that makes more sense now, I understand.

    Here is the wiki page on zfs. Seems possible with some skull sweat and some hoop jumping. Supports file sizes up to 16 exabyte, so that is certainly large enough. Good luck man!

    http://en.wikipedia.org/wiki/Zfs

  53. Been There, Fixed That by DynaSoar · · Score: 3, Interesting

    We had almost exactly the same problem. Our fMRI work was done at University of Virginia on a Linux machine. naturally you don't want to tie up a $1500/hour data collection machine doing analysis. Our data was transferred immediately to the Neurological Institute to a multiboot machine. No patient data included at this point, so no HIPPA problems. The receiving box ran Linux initially since the analysis programs from NIH (primarily AFNI) were Linux based. Patient data got added here so HIPPA became an issue. The machine had multiple hard drive bays, all of which were removable, plug-and-play drives made from a kit that provided slide-in rails and a locking mechanism, otherwise were common, commercial drives. Externals would have been easier, but the guy who devised this had a rilly rilly good reason. I remember it was good, but not what it was. Anyway, the machine could boot other OSs, prep the drives, go back to the native Linux HFS+ and transfer/translate to the , it was transferred, the drive removed, packaged, and FedEx'd to the other analysis sites at Virginia Tech, NIH, and U.Va Wise. We were strictly experimental, no direct medical treatment, and so time was not an issue. With OS X being *nix, there's not a lot of reasons to go with one over the other except for convenience when it comes to what your data collection and analysis are running under. Unless yours run fine under OS X, I'd say stick with HFS+, and of course moderate that according to whether you have to share out the data and what those people are running. I wouldn't bother with supporting Windows, as they continually find new problems to have with large files. One comparison test showed no difference in analysis results, but they did have problems with Windows choking on the data files. Their test files were only 1.5 GB. ref: J Med Dent Sci. 2004 Sep;51(3):147-54. Comparison of fMRI data analysis by SPM99 on different operating systems. PMID: 15597820. My experience agreed with their results. As I said we had little call for Macs, so we didn't run enough of that to give a good test of whether it had the same kind of problems. Bottom line, we used what we needed to according to where it was going and what they needed it to be, but for our own use it made no sense to transfer it out of the OS that collection and analysis used, HFS. The system met with the approval of the biophysicist we worked with at U.Va, and he had been a grad student under Peter Fox when the latter developed SPM. OH YEAH: the good reason. If anyone else wanted to work with us, they didn't have to dig too deeply into techie stuff either hardware or software. We could send them a removable-drive kit to install, and send them a drive with bootable Linux, AFNI and data, all plug and play. If that might be useful to you (using externals instead of removables doesn't matter here) that's probably be another vote for HFS.

    --
    "I may be synthetic, but I'm not stupid." -- Bishop 341-B
  54. Journaling unimportant - just use HFS+ or ext2 by OrangeTide · · Score: 1

    You don't need journaling for sneakernet packets. the big blobs of data are ephemeral, and users need to safely eject the drive anyways to ensure that all the data was copied.

    Journaling won't fix a file on a disk that was removed before it was completely copied over, it will have parts missing. Safely unmounted devices will also be able to skip fsck checks (up to a certain limit of days, which is configurable on many Unix filesystem, not sure about HFS+)

    MacFUSE with the ext2(or ext3) plug-in works pretty well too. Ext3 plug-in would give you that journaling feature that you find so precious.

    I'd like to see a new filesystem become the lingua franca of removable/portable media. One that is easy to implement, has an open reference implementation, is flash-friendly, supports optional extended attributes for a few OSes(name based uid/gid mappings for unix would be nice, just like tar), and generally a concise set of features so when I do embedded devel I don't have to write some crazy dancing b+tree code to access it from my bootloader. Btrfs, NILFS2, Ext4, FFS w/ soft updates, etc are all close to what I want but are too tied to their primary OS to be widely supported. UDF was a tremendous failure, it was too geared towards CD-RW and DVD media but mainly it was only weakly supported by Microsoft and Apple. No support at all would be better so a third-party could step in to maintain it, it just kind of bitrotted over at Microsoft and eventually became unusable.

    --
    “Common sense is not so common.” — Voltaire
  55. Re:HIPAA Constraints? And IT departments? by tuomoks · · Score: 1

    Sorry, a little off topic but this comment hits me every time "our IT department is reluctant to support them" - then you don't have an IT department - you have bunch of amateurs, whatever useless operators! I'm so tired of these - for real IT it doesn't matter what and how many different OS you are running - none is more or less difficult to support than any other! Yes, there are differences, one might be a little easier for some special purpose than some other but overall - they all are just a bunch of programs to run computer systems!

    Back to the subject, I have used / recommended HFS+ or, as someone said, shared filesystems over Infiniband or other fast connection. The problem is really HIPAA and other security / secret / whatever rules, regulations and laws - even today the systems don't have much good standard ways to comply - you often have to develop your own and those skills are scare! And with current education getting even more rare - back to the not so good IT!

  56. NFS+a file server... by m1ss1ontomars2k4 · · Score: 1

    That's what we use around here. Works quite well. I work with hundreds of gigabytes of MEG data myself.

  57. Find out: Phoronix Test Suite by SheeEttin · · Score: 1

    If you want to know which filesystem would perform the best, you might try just benchmarking them all with the Phoronix Test Suite. (I haven't tried it myself, but I read Phoronix' news, and there's all kind of pretty graphs.)

  58. XSAN by Anonymous Coward · · Score: 0

    Depending on the storage available.
    If you prefer to use a Promise with Apple XSan and StorNext you can use Mac, Linux and even Windows on a fiber channel system that can go up to 16 Gb/s.
    This kind of storage system can grow up to 2 PB.
    More info @:

    http://www.apple.com/xsan/features/system.html

    Tiago Rosado
    IT Manager @ Apple Distributor Portugal

    tiagoDOTrosadoATinterlogDOTpt

  59. Not ReiserFS by kevingolding2001 · · Score: 1

    I have a bunch of hard drives formatted with ReiserFS, with movies on, from ~2002. They are now basically unreadable. Don't use a format pioneered by an axe murderer.

  60. We just used FAT32 by Anonymous Coward · · Score: 0

    In our lab, in a medical school, developing anatomical atlases, we also got really big files. We had Linux, Mac OS X, and Windows systems to support, and just used FAT32 as the drives came out of the boxes. Images came off the cameras onto a Windows box controlling the stage, and onto these drives.

    We also had a Linux file server with LVM running on a RAID5 configuration for long term storage.

  61. OSX does not use fstab by SgtChaireBourne · · Score: 1

    Many filesystems support uid= and gid= options in their mount command (including HFS). Just add that to a mount script or set it up in fstab.

    OS X does not used fstab. It does leave a copy of fstab containing a message that it is not used, but no information or pointer to what it actually does use.

    --
    Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
    1. Re:OSX does not use fstab by Anonymous Coward · · Score: 0

      gp:

      Is there any good way to get Linux to mount the filesystem

    2. Re:OSX does not use fstab by Anonymous Coward · · Score: 0
      ac:

      Many filesystems support uid= and gid= options in their mount command (including HFS). Just add that to a mount script or set it up in fstab.

  62. I know MS is a bad word here but.. by Anonymous Coward · · Score: 0

    What's wrong with FAT32? If you want a filesystem that just works pretty much everywhere it's probably your best bet, and as any ACLs will be ineffective across the disparate systems I don't see why this should present you a problem.

    From experience using NTFS on external drives is a PITA and can easily corrupt cross platform, although one poster suggested this. FAT32 will just work...

    1. Re:I know MS is a bad word here but.. by ThePengwin · · Score: 1

      If they are producing files upwards of 4GB (Which they are) FAT32 is not a viable solution, as it has a max file size of 4GiB

  63. Re:HIPAA Constraints? And IT departments? by dogmatixpsych · · Score: 1

    They're a real IT department - they are very good - they are just under-staffed due to budget constraints (that's one problem with being at a public university). We could let them manage our systems (they do whatever we ask them to do) but we actually wanted to manage our own computers and IT, being reluctant in the first place to add Macs to the infrastructure, is fine with us doing (most of) the managing.

    Thanks for the comment. I've tested the HFS+ set up and it seems to be working very well.

  64. Re:UDF IS ACTUALLY A SOLUTION by pknoll · · Score: 1

    You can create the filesystem from OS X, too:

    newfs_udf -v /dev/diskX

    I'd suggest the man page for newfs_udf for the many options.

  65. PACS by chihowa · · Score: 1

    I'll second that. There are plenty of free PACS servers available, and you could also ask your MRI vendor for a recommendation (or buy their super expensive version). I've used Osirix before, which I like as a viewer but gets slow when you start building huge libraries of data.

    Because I hate when people respond to questions by saying you're asking the wrong question, I'd suggest sticking with HFS+ for file transfers. I don't think you have journalling on Linux, yet, but otherwise Linux HFS+ support is very good. For filling a drive and walking it down the hall, journalling is a little unnecessary anyway. Journalling comes into play more when you're working with the data on a drive. For loading/unloading, journalling will just slow the copy process down a little. If you're worried about the integrity of your copy, do a verify after the copy... journalling won't help you there anyway.

    --
    If you want a vision of the future, imagine a youtube comments section scrolling - forever.
  66. Re:UDF IS ACTUALLY A SOLUTION by Ant+P. · · Score: 1

    Whole disk? I've been using partitioned SD cards with UDF for ages now, what's wrong with doing that?

  67. Re:UDF IS ACTUALLY A SOLUTION by Psymin · · Score: 1

    This is awesome! Thank you so much for helping me ditch FAT32 :)

    If anyone knows how to convince xp to write to a device like that (via a 3rd party driver or whatever) I'm all ears!

  68. case sensitive HFS+ by Anonymous Coward · · Score: 0

    yeah, tried that once, lotsa stuff breaks, mostly windozw-heritage installs:-(

  69. I actually tried, after your suggestion by Ilgaz · · Score: 1

    After reading your post, I actually formatted the external drive to UDF 2.51 considering it will be used on Win Vista and OS X latest (SL).

    It is a perfect filesystem in theory. In reality, it is either Vista or OS X to blame, it didn't really work as theorized. OS X Disk Utility kept whining, neither OS used the excellent features (store native metadata etc.) offered and lack of "fsck" utility killed it for me.

    Now, I know such a excellent option exist without any lack of features and can store native stuff, I got really pissed.

    Thanks to "cold war" between both companies, Apple and Microsoft, I don't know who to blame. I went back to old fashion "divide disk to 2, use ntfs and hfs+" method.

    BTW; I am not the "MR" guy, I am the guy you replied to.