Slashdot Mirror


32 GB Flash Storage Drive Announced

Audrius writes to tell us TG Daily is reporting that Samsung has just announced a new 32 GB Flash storage device. The aim of this new solid state disk (SSD) drive is to completely replace the traditional hard drives in many laptops on the market. Some of the advantages offered are the 1.8" form factor, read speeds more than twice that of a normal hard drive, and the promise of 95% less power use.

381 comments

  1. Digital Camcorders by Shadow+Wrought · · Score: 5, Interesting

    I could see this having a pretty big impact on digital video cameras, too. No moving parts to break while you're running around with a handheld. Very cool!

    --
    If brevity is the soul of wit, then how does one explain Twitter?
    1. Re:Digital Camcorders by jessecurry · · Score: 2, Insightful

      That actually would be pretty nice, although I think that the price is still well above that of magnetic tape; maybe it would be a little more useful in a professional setting where the video could be pulled onto an editing station and then erased from the original flash media.

      --
      Those who know, do not speak. Those who speak, do not know. ~Lao Tzu
    2. Re:Digital Camcorders by firl · · Score: 1, Interesting

      Not only that, think about an ipod that doesn't have the hard drive problems that it does today. or imagine a raid of these in a server for either a SAN or a Database, where the data i/o would surpass 1 gig per second, granted if they make it go beyond the normal pci bus. I personally am looking forward to this

    3. Re:Digital Camcorders by Eightyford · · Score: 1

      I could see this having a pretty big impact on digital video cameras, too. No moving parts to break while you're running around with a handheld. Very cool!

      Definitely. I always thought those dvdburner camcorders were a bad idea.

    4. Re:Digital Camcorders by xerid · · Score: 1

      Panasonic already has one, but the P2 cards are over priced IMO:

      http://www.panasonic.com/business/provideo/hvx200/ index.asp

    5. Re:Digital Camcorders by cayenne8 · · Score: 1
      Yes, but, don't 'flash drives' suffer from a more limited number of times you can read/write to them than a regular HD?

      That wouldn't be too good for a camera or a laptop...

      --
      Light travels faster than sound. This is why some people appear bright until you hear them speak.........
    6. Re:Digital Camcorders by rwven · · Score: 1

      Supposedly... From what i've heard, the actual limit is so high that it might as well be the same as a reg. HD... They "rate" them for a certain amount mainly for warranty purposes and they can (again: supposedly) do way beyond that.

      Anyone with "yes or no" power around?

    7. Re:Digital Camcorders by drgonzo59 · · Score: 1

      I am not sure about hard drives but with flash, each bit has a limited # of times it can be read/written. So if you have an ext2 file system that writes the access time back to the disk every time the file is read, then there will be certain a certain set of bits that will be disproportionately accessed and thus would be the first to fail. I had Zaurus Linux handheld computer and when I made sure to add the noatime option when mounting the partitions. But of course, if the limit can be fairly high, then it won't be a problem....

    8. Re:Digital Camcorders by Vlad2.0 · · Score: 1

      Yes. Eventually the transistors used to store data will be physically damaged and unusable. This number is very hardware dependent. Depending on the technology it can be from as little as ~100 to ~10^5 times. Flash memory should be at least 10^4 to 10^5 times...if anyone has more accurate data, I'd be interested in knowing it.

    9. Re:Digital Camcorders by networkBoy · · Score: 1

      Yes they are limited.
      The mean cycle count is 150K+ on the devices I am most experienced with though the warrenty is 100K cycles.
      -nB

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    10. Re:Digital Camcorders by HardCase · · Score: 2, Informative

      Yes, but, don't 'flash drives' suffer from a more limited number of times you can read/write to them than a regular HD?

      The typical number tossed around for NAND erase cycles is 100,000. You can read as often as you like, but to write data, you have to erase a block of data first, 132KB on the devices that I design with.

      Of course, those are the data sheet numbers - that is what the manufacturer guarantees. Reality is usually quite a bit better. And it wouldn't surprise me if Samsung and others had some much higher performance flash memory in the pipeline.

      -h-

    11. Re:Digital Camcorders by petermgreen · · Score: 1

      afaict modern disk like flash has controller chips that perform wear leveling functions to basically eliminate this issue.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    12. Re:Digital Camcorders by Animaether · · Score: 1

      Of course it's too expensive... consider what it is;
      "The PCMCIA card is actually an array of secure digital cards designed to work swiftly and in harmony to record large amounts of data"
      http://www.camcorderinfo.com/content/Panasonics-P2 -Format-Explained.htm

      It's like a Flash RAID stuck in a PCMCIA housing. Except you pay a premium for it. Which they can charge as there's no real aternative (other than tapes and HDDs)

    13. Re:Digital Camcorders by SphericalCrusher · · Score: 1

      Yeah, and for Digital Cameras too. I think this would make an interesting gift for my girlfriend's Sony camera... if only it were available on a card. Hah.

      --
      "Instant gratification takes too long." - Carrie Fisher
    14. Re:Digital Camcorders by sgidude · · Score: 1

      Panasonic's P2 uses SD cards in a PCMCIA frame with raid for exactly this.

    15. Re:Digital Camcorders by dnoyeb · · Score: 1

      No its not that high. And it also degrades faster when written at high or low temperatures.

    16. Re:Digital Camcorders by Anonymous Coward · · Score: 1, Insightful
      Personally, I don't think he's a twat.

      Hell, I think you're a twat for having such an inane irritation such as that. Twat.

    17. Re:Digital Camcorders by TheBeowulf · · Score: 1

      No moving parts... except for the lens zoom and focus!

      A completely solid state camcorder can't work without optical zoom.

    18. Re:Digital Camcorders by Anonymous Coward · · Score: 0

      Almost any "mobile" piece of equipment would be able to benefit. I know I am doing the anonymous coward thing, but one thing that comes to mind is Orgami. Despite the M$ platform, this could allow for you to have the main portion of your OS and some subset of your data stored on this, then you can just swap this card between your desktop and mobile device (PDA, laptop, etc (Orgami?)). You might keep most of multimedia on your desktop on your 500GB+ drive and then just synce your playlists or your current weeks worth of TIVO'ed movies. Plus, if you wanted to get fancy, the OS could know what files were on the Flash card and what was on the hard drive and then if the hard drive wasn't accessible then it would try to do a mount of the remote drive...like when your desktop has the Flash drive removed, it acts only as a fileserver for when the mobile device want's to access it.

      Its pretty dern crazy, but if someone could make it seamless then it could be really powerful.

    19. Re:Digital Camcorders by nmg196 · · Score: 1

      > From what i've heard, the actual limit is so high that it might as well be the same as a reg. HD

      Not in my experience. I've been using hardware firewalls which are basically a mini PC motherboard but with an SSD device which emulates an IDE hard disk. All three of these devices suffered a failure of the SSD within a year. Since I replaced the SSD with a REAL laptop hard disk, the device has been running fine for nearly 3 years. I've heard that some of the non-persistant devices can take more read/write cycles, but they are non-persistant and therefore require a hard disk to backup to anyway - negating most of the point of them.

    20. Re:Digital Camcorders by vertinox · · Score: 1

      A completely solid state camcorder can't work without optical zoom.

      No real camera man uses the built in auto-focus. ;)

      --
      "I am the king of the Romans, and am superior to rules of grammar!"
      -Sigismund, Holy Roman Emperor (1368-1437)
  2. Sounds Really Promising by Anonymous Coward · · Score: 0

    But, so do most politicians.

    Results is what I want. Deliver it to market at a reasonable price and I'm onboard. Else it's just more worthless talk.

    1. Re:Sounds Really Promising by Anonymous Coward · · Score: 0

      I am sure the CEO of Sony is taking your words to heart. :)

  3. Interesting .... by GoodOmens · · Score: 4, Informative

    This will only work if they can get the prices of flash down.
    $50.00~70.00 per gb is still nothing in comparison to $0.40~$0.80 you can get on hard drives.

    1. Re:Interesting .... by Anonymous Coward · · Score: 2, Informative

      Seagate 8 GB CompactFlash Photo HD $149.99 shipped free, Mar 20

      amazon.com has the new Seagate 8GB CompactFlash Photo HD ST68022C-RK for a low $149.99. No rebates. Free shipping. Tax in KS, ND, WA.

      4GB $74.09 shipped free.

      from techbargains.com

      $18.5 / GB -- and who here doesn't remember when a 100MB hard drive was $300+?

    2. Re:Interesting .... by TERdON · · Score: 1, Informative

      Those are real hard drives, with flash memory interfaces. Good try though, and $6400 probably is too high anyway (4 GB CF flash cards can be found for approximately $300).

      --
      I have a really elegant proof for Fermat's last theorem. If this sig was only a bit longer...
    3. Re:Interesting .... by NitsujTPU · · Score: 4, Funny

      Oh, sure, but when you consider the power savings mentioned in the summary, those prices really start to pale! It costs me nearly a quarter of a million dollars a year to charge my Dell Inspiron 9100, with an old fashioned hard drive.

    4. Re:Interesting .... by GoodOmens · · Score: 3, Interesting
      Ehh so my math is slightly off. Its still roughly 37x the cost of a hard drive.

      Anyways you are right though. I can see solid state drives taking over hard drives in the future. The less moving parts the better.

      All I was trying to point out was its to early now for widespread adoption.

    5. Re:Interesting .... by moosesocks · · Score: 1

      Hard-drive prices aren't falling nearly as fast as they were a few years ago.

      The price of flash, on the other hand, is in a total free-fall. We're paying a lot less than $50/gb right now. 10 years ago we were paying close to $50/mb.

      You also have the issue that laptop drives tend to be pretty small due to space and heat restrictions. The biggest laptop HD I've seen is 120gb, and drives that big are rare and expensive. If you could get a speedy solid-state 60gb drive for about $300, the market for it would be huge

      --
      -- If you try to fail and succeed, which have you done? - Uli's moose
    6. Re:Interesting .... by ucblockhead · · Score: 1

      It isn't all about money. Flash has advantages over hard drives, and some people will likely pay for those advantages. (Just look at the number of iPod Nanos that sell.) I suspect a fair number of people would buy a laptop with flash memory despite the premium as it'd be lighter and have much better battery life.

      --
      The cake is a pie
    7. Re:Interesting .... by gstoddart · · Score: 5, Informative
      Those are real hard drives, with flash memory interfaces. Good try though, and $6400 probably is too high anyway (4 GB CF flash cards can be found for approximately $300).

      Hmmm ... are you sure about that? They call it a Solid State Disk and say it's based on NAND Flash memory.

      The article certainly sounds like it's not using any spinning-platter/read-write heads technology -- that would not really be solid-state. That seems to be how it uses less energy and makes no noise.

      To me, this doesn't sound like a "hard drive", but a big whack of Flash memory which is treated like a hard-drive.

      The $6400 figure comes from the article:
      One of the few 32 GB Flash disks on the market is currently sold by Silicon Systems: The device comes in a PCMCIA form-factor and is priced around $6400.

      So, it's not like the posted pulled the number out of thin air.

      If it's got no moving parts, it's not what we would traditionally call a hard-drive.
      --
      Lost at C:>. Found at C.
    8. Re:Interesting .... by cliveholloway · · Score: 3, Insightful

      $300? I just bought one for $160 on eBay. I think it must be a while since you last looked at prices ;-)

      --
      -- Trinity in high heels carrying a whip: The donimatrix - there is no spoonerism
    9. Re:Interesting .... by fbg111 · · Score: 1

      If 4GB is $300, then 32GB is $2400, not $6400. Not that it particularly matters, that's still more than most desktop PCs and lowend servers anyway.

      --
      Flying is easy, just throw yourself at the ground and miss. -Douglas Adams
    10. Re:Interesting .... by God'sDuck · · Score: 1
      Hmmm ... are you sure about that? They call it a Solid State Disk and say it's based on NAND Flash memory.
      The 8 GB model he refers to solid-states at 3600 rpm:
      http://www.seagate.com/products/retail/flash
    11. Re:Interesting .... by TERdON · · Score: 1

      the OP that I was responding to wasn't saying anything about SSD. I was referring to that and nothing else. As you correctly point out, TFA is about SSD disks, why I wanted to point out that there are hard drives with compact flash form factor (like the ones the OP managed to find)

      --
      I have a really elegant proof for Fermat's last theorem. If this sig was only a bit longer...
    12. Re:Interesting .... by TERdON · · Score: 1

      Thanks, I just took my favorite online store and checked its prices. I happened to get the prices of the extra-expensive extra-high-speed ones. They have some for approximately $200 as well.

      --
      I have a really elegant proof for Fermat's last theorem. If this sig was only a bit longer...
    13. Re:Interesting .... by petermgreen · · Score: 1

      I can see solid state drives taking over hard drives in the future
      i don't see that happening unless increases in hdd size and software bloat stop. If you only wan't to run 486 era software (windows 95 office 97 some small games etc) theres nothing stopping you running off a CF card now without paying a huge ammount. but buying enough to support current bloatware is prohibitively expensive. I don't see how this is likely to change sure the cost per megabyte of flash will go down but software bloat will probablly keep going up.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    14. Re:Interesting .... by gnuyarlathotep · · Score: 1

      Those are real hard drives, with flash memory interfaces. Good try though, and $6400 probably is too high anyway (4 GB CF flash cards can be found for approximately $300).

      You sure about that? My 4GB Nano was $249 months ago.

    15. Re:Interesting .... by Anonymous Coward · · Score: 0

      "who here doesn't remember when a 100MB hard drive was $300+"

      IIRC my first external CMS SCSI 173M hd was around $1k(+/- $300) around 1989/90... (attached to a IIcx)
      (sounded like the flight deck of an aircraft carrier too...)
      (... the enclosure/PS are still up and running, but the drive itself died about 10 years ago, and was replace with a 230M drive...)

    16. Re:Interesting .... by Hadlock · · Score: 2, Interesting

      I know you're kidding, but I'm excited about keeping my rapidly aging 2001 powerbook on the road by upgrading it's 20gb traditional hard drive with a 32gb flash drive in a year's time. When the battery was new, I could get 5 hours of web surfing time in with the HD spun down. I suspect a flash drive would use even less energy than a hard drive in idle mode.
       
      The other thing people haven't mentioned, is that many laptops use 4200 or 5400rpm drives to conserve power, which often become the limiting factor for speed on the laptop. Currently I use a 7200rpm external drive over firewire, and I picked up about a 15% increase in "percieved speed" according to my Hadlock-meter. A flash drive would give me the same sort of performance on the road, without the need for a bulky external drive + wall wart.

      --
      moox. for a new generation.
    17. Re:Interesting .... by Muramasa · · Score: 0

      You're forgetting that the one you bought on eBay is stolen. The retail price is probably closer to $300.

    18. Re:Interesting .... by Anonymous Coward · · Score: 0

      I remember 10 megabytes for 5 figures cash in early 60's money. My dad was a mainframe guy and brought it (a piece of the drive anyway) home after it broke. He said something like 10 grand for one of them new, something like that, long time ago now. About three feet across IIRC maybe a tad smaller, big enough though and heavy. He built a patio coffee table out of it. It was a large plexiglass platter. How it worked I can't tell ya.

      My personal earliest computer only had a single sided floppy, no hard drive at all in it. A mac512. Still have it.

    19. Re:Interesting .... by DynamiteNeon · · Score: 3, Informative

      I know some people modded that funny, but I have an Inspiron 9100 also, and he's not joking. Nice laptop, but it'll suck faster than a cheap hooker.

    20. Re:Interesting .... by TubeSteak · · Score: 1
      Anyways you are right though. I can see solid state drives taking over hard drives in the future.
      Maybe... maybe not. Hard drive platters are made of glass or aluminum, while flash memory is made of silicon.

      No one is talking about the manufacturing end of the equation. Will there be enough of the raw materials?

      I remember reading about solar cells & one of the main reasons costs aren't dropping, is because the high demand has jacked up silicon prices. In a few years, more manufacturing capacity will come online and prices will drop, but what if flash memory starts competing for these limited resources?

      In the long term, sure, the technology will come to market. But 'short' term prices might stifle the technology long enough for something else to roll around. I personally can't wait for holocubes.
      --
      [Fuck Beta]
      o0t!
    21. Re:Interesting .... by The+Mayor · · Score: 1

      Oh, man, I think I had better buy some silicon dioxide futures then. Or, I suppose, we could always just go to a desert & pick up some sand...

      Kidding aside, manufacturing cost curves for solid state follow the First Corrolary to Moore's Law--that the price per transistor halves every 18 months (Moore's Law states that the # of transisters doubles every 18 months, but the corrolary also proves true). According to this article, the period for photovoltaic cells is about 4-6 years. Hard disk drives seem to be following a period of 2-5 years, with a greater deal of variance than with solid state. It is inevitable that solid state will be cheaper than hard disk drives at some point.

      The same holds true for LCD panel monitors. Some day, wall-sized LCD panel monitors will be only slightly more expensive than wallpaper, and we'll be living in a Total Recall world.

      Back about 15 years ago, I calculated the crossover point for solid state vs. hard disks as occuring in 2008. I don't think I'll be too far off...maybe 2010?

      --
      --Be human.
    22. Re:Interesting .... by adarn · · Score: 1

      My first harddrive was the Applied Engineering Vulcan for the Apple IIgs with an astounding 105 MB of storage. I dont know the manufacturer of the actual drive, the Vulcan was a self contained unit that included the drive, a proprietary (scsi i think) interface and a higher rated power supply (the drive and the power supply were in the same black case, which replaced the old power supply). It was $1300 in I think 1987 or 1988. I really wish I could have all the money I've spent on computers in front of me right now. I'd move to europe.

      Adarn

    23. Re:Interesting .... by v1 · · Score: 1

      I picked up a 4gb cruzer mini a few months back for $260, and I'd say it was worth every penny. I've worn the finish off the plastic. It gets used several times a day on average where I work.

      I'm waiting for the 8gb drives to come out. This one's already got less than a gig free on it. They usually double the capacity every 4-6 months, so they're due to double again real soon.

      The cruzer mini is no hippo either, like those ridiculous round seagate monstrosities or the "blocks every adjacent port" jetflashes. There are two things I'd change if I could though... the ability to function if plugged into an unpowered hub (like a keyboard!) and a write protect switch. (if i was doing av work on PCs the lack of write protect alone would be a show stopper)

      --
      I work for the Department of Redundancy Department.
    24. Re:Interesting .... by Mignon · · Score: 1
      I suspect a flash drive would use even less energy than a hard drive in idle mode.

      Yes, but I think it's the backlight and/or CPU that draw the lion's share of the juice. I looked into this idea a while back. Sorry, don't recall the numbers, but they should be easy to find.

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

      If you keep it on your lap (while plugged in) the Dell Perspiron is also hotter than the afore mentioned cheap hooker.

    26. Re:Interesting .... by miro+f · · Score: 1

      FTFA: "While the SSD's capacity of 32 GB cannot compete with traditional hard drives that currently offers up to 80 GB space"

      clearly the article is many years old. I'm sure the price of the flash SSD has decreased in the meantime ;)

      --
      being vague is almost as cool as doing that other thing...
    27. Re:Interesting .... by Anonymous Coward · · Score: 0

      I'm sure TFA is refering to the capacity of 1.8" drives. Not hard drives in general.

    28. Re:Interesting .... by Anonymous Coward · · Score: 0

      Yeah, big advantages. You want to know how this is going to be up-sold? No, not battery life, though that will come in.

      Fear. Drop your Old-School-HardDrive laptop, and you could lose not just the data you were working on, but EVERYTHING ELSE TOO! (rising pitch, raised eyebrows).

      And it actually has the benifit of being true. No more worrying about moving your PC while it is shutting down or booting. No more worries about wiping a drive if the laptop gets dropped.

      A brave new world, indeed.

    29. Re:Interesting .... by nmg196 · · Score: 1

      > Those are real hard drives, with flash memory interfaces

      Er no. They're clearly stated as SSD devices. ie solid state. RTFA.

      What your describing is IBM MicroDrives, which obviously don't cost $6400 or nobody would buy them.

    30. Re:Interesting .... by richarnd · · Score: 1

      "Maybe... maybe not. Hard drive platters are made of glass or aluminum, while flash memory is made of silicon" Erm...what do you think glass is, primarily? This is like saying demand for ice won't affect water prices.

    31. Re:Interesting .... by TERdON · · Score: 1

      RTFP (Parent). I wasn't talking about the article goddammit.

      --
      I have a really elegant proof for Fermat's last theorem. If this sig was only a bit longer...
    32. Re:Interesting .... by Anonymous Coward · · Score: 0
    33. Re:Interesting .... by nmg196 · · Score: 1

      Sorry - my mistake.

    34. Re:Interesting .... by NitsujTPU · · Score: 1

      Eh, not the 9100's. There is a lot of infrastructure between you and the real heat.

      A D600 will burn you quite nicely though.

  4. Data Integrity by jessecurry · · Score: 4, Interesting

    Will this still be useful for critical applications? What's the current failure rate of flash memory?

    --
    Those who know, do not speak. Those who speak, do not know. ~Lao Tzu
    1. Re:Data Integrity by Zemrec · · Score: 1

      I'm far from an expert, but the failure rate for modern flash is much much better than for hard disks. Something like millions of write operations per bit before failure.

      And correct me if I'm wrong, but aren't lots of enterprise-class servers, routers, switches, etc. booted from PC-card style flash drives?

    2. Re:Data Integrity by temojen · · Score: 4, Informative

      Flash memory that works has a much longer MTBF than hard drives, but each cell fails at approximately 10000 writes. HDDs fail randomly, Flash fails predictably, so this can be a good thing. Just make sure your filesystem rarely does or needs defragging, and does not log every read.

    3. Re:Data Integrity by Jordan+Catalano · · Score: 1

      I'd be more worried about bits getting "stuck" that reveal critical data that can no longer be securely erased.

    4. Re:Data Integrity by From+A+Far+Away+Land · · Score: 1

      "critical applications"

      Are you wondering too if the Porn collection will be safe? And does it suffer from wear if the same file is read over and over again?

      Sorry, it's a standard Slashdot joke when storage is mentioned. Just be glad I didn't ask how many Libraries of Congress it can hold.

    5. Re:Data Integrity by poopdeville · · Score: 2, Funny

      The solution to that issue is the same as it would be for disk based drives: thermite.

      --
      After all, I am strangely colored.
    6. Re:Data Integrity by nbert · · Score: 4, Insightful
      Just make sure your filesystem rarely does or needs defragging, and does not log every read.
      On a flash drive it's not really important into how many segments a file is split or where they are located since there's no head spinning back and forth. So there's only a problem if your fs does defraging automatically, but it's quite easy to switch this off (at least for developers)
      Guess we have to reconsider some habits we've got accustomed to if traditional hds are replaced.
    7. Re:Data Integrity by ucblockhead · · Score: 1

      "Defragging" is a hard drive concept. It makes no sense with a random access storage system.

      --
      The cake is a pie
    8. Re:Data Integrity by temojen · · Score: 1

      Assuming your filesystem keeps parts of only a single file in each block. Not all filesystems do.

    9. Re:Data Integrity by TACNailed · · Score: 1

      I'm sure you can share your reliable source on that 10,000 writes number with us. That number is an order of magnitude less that what I have heard.

    10. Re:Data Integrity by ucblockhead · · Score: 1

      It doesn't matter. With a random-access storage mechanism, there's no cost if the blocks that are in a file are not sequential.

      --
      The cake is a pie
    11. Re:Data Integrity by tuxguy_ga · · Score: 1

      It makes for an interesting virus / worm concept to kill a drive. Just trigger constant flash writes...

    12. Re:Data Integrity by Anonymous Coward · · Score: 0

      That would be my concern as well, the number of writes that each cell of NAND flash can support, esp. when used with a full VM enabled OS. I note that the article doesn't even mention this possible limitation, so it is possible, one might infer, that they have managed to increased the # of writes possible astronomically, although I would have expected that to have been the highlight of the article in that case.

    13. Re:Data Integrity by Danga · · Score: 1

      Reread the guy who responded to you. Yes, a highly fragmented file on a flash chip won't have the reading slowdown compared to the same file being highly fragmented on a hard disk due to the flash drive not having any moving parts. However, there is a cost if the memory blocks are shared between files since the disk could be better utilized keeping all files contents within the minimum amount of blocks necessary/possible. The advantage to defragmenting a flash drive that does share blocks between files is not faster reading speed, but better memory utilization. Who here doesn't have a need/want to have as much free space available as possible?

      --
      Hey, there is only one Return and it's not of the King, it's of the Jedi.
    14. Re:Data Integrity by jfw25 · · Score: 1
      How reliable a source would you consider http://www.sandisk.com/Assets/File/OEM/WhitePapers AndBrochures/CompactFlash/oem-product-line-bro.pdf ? They specify > 2,000,000 program cycles per block for their Industrial PC card Flash drives and > 300,000 cycles per block for Standard Grade PC card drives (which certainly handily beats the 10,000 figure you're objecting to). Granted, this brochure doesn't list write cycle specifications for the cheap consumer-grade packages, but I doubt they're as bad as ~10,000 cycles.

      And from having been intimately involved in the persistent-storage subsystem of a telecommunications product using those Flash cards, that program-cycle specification is per real block, not taking into account the wear-levelling and sparing feature. As I recall (from very dim memory) the PC card devices had about 15% spare capacity for replacement blocks. (That brochure PDF doesn't include that figure, though, so I can't be sure of that figure.)

      However, these specifications are for their newest products; four years ago they were nowhere near that good. With a much older revision of this same Flash product (and with, ahem, some highly suboptimal write behavior) we were cooking some Flash disks in mere weeks in the lab...

    15. Re:Data Integrity by Maximum+Prophet · · Score: 1

      Swapping to a flash card would kill it fast.

      --
      All ideas^H^H^H^H^Hprocesses in this post are Patent Pending. (as well as the process of patenting all postings)
    16. Re:Data Integrity by jessecurry · · Score: 1

      Is there a critical application other than porn? I don't care about my financial records or personal documents as long as I can see boobies :D

      --
      Those who know, do not speak. Those who speak, do not know. ~Lao Tzu
    17. Re:Data Integrity by Shanep · · Score: 1

      Flash memory that works has a much longer MTBF than hard drives, but each cell fails at approximately 10000 writes.

      The 10,000 number is what the really old flash devices were rated at. I've heard various numbers, which depend on the flash memory type and layering structure. I've seen numbers from the hundreds of thousands to millions. Cards which are large but slow, tend to use multiple layers of flash and are not as durable. Cards which are smaller but fast (the "Pro" and industrial cards), tend to use a single layer and last the longest.

      I can't seem to find the document which details this at the moment.

      I hope they one day manage to make flash as durable as DRAM, but still retain the non-volatile nature. I'd love to have solid state storage in my laptop (or everywhere for that matter).

      --
      War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
    18. Re:Data Integrity by temojen · · Score: 1

      Datasheet for AT29C256. It's an old design though.

    19. Re:Data Integrity by Fweeky · · Score: 1

      "I hope they one day manage to make flash as durable as DRAM, but still retain the non-volatile nature."

      To hell with flash, we want MRAM.

    20. Re:Data Integrity by Shanep · · Score: 1

      To hell with flash, we want MRAM.

      You have reminded me of FRAM's. Touted in the late 80's to be the future of memory devices. Fast and non-volatile.

      But MRAM sounds awesome. Non-volatile, does not degrade when written to, writes are not slower than reads and almost as fast as SRAM!

      I had high hopes for FRAM but it didn't seem to come out of the labs. The Wiki for it states how much better FRAM is over Flash, but I've been waiting something like 18 years for FRAM. Hopefully MRAM will be used in a big way.

      --
      War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
  5. Reliability? by smoor · · Score: 3, Interesting

    It seems like a nice way to go (solid state). I wonder what the life of a unit like this would be. Flash drives might be droppable, but what else can kill them? Somehow I feel better imagining that my stuff is magnetically etched into a platter... I guess I'm just old...

    1. Re:Reliability? by manifoldronin · · Score: 2, Insightful

      Somehow I feel better seeing that my stuff is physically etched into a piece of paper... I guess I am just old... 8-)

      --
      Tyranny isn't the worst enemy of a democracy. Cynicism is.
    2. Re:Reliability? by ChrisGilliard · · Score: 1

      If you were using this with the GDrive (Google's online unlimited storage thingy), maybe reliability is not so important because you could have backups made in real time to the GDrive. Also, you could do some sort of caching as well. Imagine a 32 gB local cache and storing the rest of your data on the network so that you can really have super fast unlimited memory capacity.

      --
      No Sigs!
    3. Re:Reliability? by Anonymous Coward · · Score: 0

      My question is one of security. If the memory can't be overwritten many times without wearing out the component, then how can you do a iterative secure wipe of data when it's deleted. Wouldn't this be defeated by both the lack of allowable writes and the components own measures to extend the life of the component?

    4. Re:Reliability? by diskis · · Score: 1

      When you forget your piece of paper in your pocket and wash your pants, what's left?
      When I forget my SD card in my pocket it survives. I have a card here that's gone through a washing machine 3 times. Still works.

      I guess you have to carve your stuff in stone tablets to get something you can see and which will also last. :)

    5. Re:Reliability? by Mr+Sausage · · Score: 1

      Well my old Byetstor 256MB has been through the wash several times (somewhere between 8 and 10 according to my wife) and is still going. Try doing that with a conventional HDD.

      --
      "Hello, I'm a british person"
    6. Re:Reliability? by Anonymous Coward · · Score: 0

      You don't have too put in nearly that much effort. Just modify a printer(the kind that can write on CD's) to write on pieces of metal with chemicals that are reactive with that specific metal.

    7. Re:Reliability? by jarom · · Score: 1

      Isn't it the number of erase cycles that is the problem?

      --
      This signature is far too complex to have been created by chance.
    8. Re:Reliability? by Mr+Sausage · · Score: 1

      No it's the spin cycle you need to watch out for......

      --
      "Hello, I'm a british person"
    9. Re:Reliability? by m50d · · Score: 1
      Somehow I feel better imagining that my stuff is magnetically etched into a platter...

      I certainly don't. It's too easy to kill magnetic stuff - magnets (duh) but also airport security, anything electrical,.... This is like the move from cassettes to CDs.

      --
      I am trolling
  6. Welcome to the future by bcoff12 · · Score: 1

    the future is now....

    1. Re:Welcome to the future by Lispy · · Score: 2, Insightful

      hm. Where I come from the future seems to be always 5 years in the future. ;-/

    2. Re:Welcome to the future by Anonymous Coward · · Score: 0

      That's okay. I'm posting this from 10 years in the future.

    3. Re:Welcome to the future by manual_overide · · Score: 1

      At least you are not in Cincinnati like me. Here, the present is 5 years into the future (at least Mark Twain seems to think so)

      --
      If bad puns were like deli meat, this would be the wurst
  7. What about the limited number of writes? by jay2003 · · Score: 4, Insightful

    I have the understanding that flash memory has a finite number of writes and that conventional filesystems with their update of metadata even on file read could essentially wear out a flash drive quickly if it was used as the main disk drive (as opposed to digital camera use or the like where access is comparably infrequent)

    1. Re:What about the limited number of writes? by GoodOmens · · Score: 3, Informative
      Hard drives also have the same limitation (You can only change the poll of a bit on the hard drive a certain amout of times). Its just you will never reach it before the mechanics of the drive fail.

      Its just a matter of time for flash.

    2. Re:What about the limited number of writes? by Anonymous Coward · · Score: 0

      Yeah, I thought so too - this is why portable Fx has a few changes to the settings to reduce the number of disk writes. Anyone care to clarify?

    3. Re:What about the limited number of writes? by LehiNephi · · Score: 3, Informative

      The current number-of-writes for flash is somewhere in the 100,000-1,000,000 write cycle range. That's a lot of writes. Also, keep in mind that all flash chip controllers include logic that performs "write-leveling". This means that a specific chunk of data will 'jump' from one area of the memory to another in order to prevent one area from being worn out. Add to that the fact that flash chips contain some extra capacity to compensate for bad blocks.

      With a careful configuration of Windows (no page file, no IE cache, no temporary files, use a RAM disk), this is certainly viable. In the absense of music/movie collections and monster games, even the 32GB size isn't that restrictive.

      --
      Help find a cure for cancer. Join the [H]orde
    4. Re:What about the limited number of writes? by Anonymous Coward · · Score: 0
      pole



      just thought you'd want to know

      cheers

    5. Re:What about the limited number of writes? by kebes · · Score: 5, Informative

      This is certainly a valid worry. As I understand it, however, modern flash memories have more or less dealt with this problem, because:
      (1) The number of rewrites is now quite large (hundreds of thousands?)
      (2) The writing-to-disk software/hardware implements "load balancing." If you rewrite the same file 1000 times, it won't use the same exact block on the flash disk for each of those writes. Instead it will move from block to block with various writes/deletes/modify actions. This, coupled with some "slack" (the actual disk size is a little bit bigger than the "useable" disk size) allows for the wear to be distributed over the whole device.
      (3) The system uses conventional error-correction and flagging of bad blocks.

      As another poster pointed out, magnetic hard disks also have a limited number of rewrite cycles. But in practical terms we usually don't reach this limit. For critical applications I imagine you'd use a RAID of flash disks just like a RAID of magnetic drives.

    6. Re:What about the limited number of writes? by soft_guy · · Score: 1

      I could certainly imagine using this on a laptop especially. What if you could make a laptop with this as the main drive and a iPod sized hard drive as a secondary drive that you put "media" (movies, music, etc.) on?

      --
      Avoid Missing Ball for High Score
    7. Re:What about the limited number of writes? by MindStalker · · Score: 1

      With a careful configuration of Windows (no page file, no IE cache, no temporary files, use a RAM disk), this is certainly viable. In the absense of music/movie collections and monster games, even the 32GB size isn't that restrictive.
      Something like this will most likly be used to store the OS and main programs essentially being a read only drive, while a second drive can be used to store data and temporary files. I really don't see the advantage here unless the flash memory is much faster though from my experience its slower, I guess they are using a new technology??

    8. Re:What about the limited number of writes? by sgbett · · Score: 0

      I'll say!

      I'm getting by just fine on my Dell L400 and its 9 gig drive. Which not only lets me surf from the sofa but also runs a full webdev environment (LAMP) quite comfortably....

      # df
      Filesystem 1K-blocks Used Available Use% Mounted on
      /dev/hda3 9285280 8354664 930616 90% /

      Not much space for pr0n though :)

      --
      Invaders must die
    9. Re:What about the limited number of writes? by Anonymous Coward · · Score: 0

      Typicall flash drives are at the mercy of their interface. For example you plug a USB2 flash drive into a USB1 port (it will fit) but it will run at USB1 speeds. However if you get it to use USB2 it is MUCH faster.

      Howwwwwwwwwwever there are crummy flash chips out there. Not all flash chips are made the same way. Some have awsome read speeds yet very poor write speeds. Some have pretty good write and average read. Its all about trade off.

      This definatly will have an interesting effect in the static device market, such as iPods or Kiosks. Where the data does not change much but you need gobs of data.

      I too was worried about number of writes with a device I was working on. I have yet to see an RMA with a 'defective' drive. We write TONS of data to the things. Course this is out of a sample size of thousands.

    10. Re:What about the limited number of writes? by TopSpin · · Score: 2, Insightful

      The writing-to-disk software/hardware implements "load balancing."

      This will be further enhanced with small, battery backed RAM write cache integral to the device. BBWC is commonplace in storage. Flash storage (eventually it will occur to us that emulating disks isn't useful) will just scale it down to a few hundred kilobytes + tiny battery and some large percentage of writes direct to Flash will not occur. Between the write cache and write balancing you'll get many years of use, and failure predicted by a simple progress bar as the device approaches its write limit.

      This will, of course, take about a decade of hashing around with new "standards", including excellent proprietary solutions from Apple that won't go anywhere due to royalties, various bad reimplementations from everyone else that will complicate the market and slow adoption, etc.

      Enjoy.

      --
      Lurking at the bottom of the gravity well, getting old
    11. Re:What about the limited number of writes? by teslar · · Score: 2, Interesting
      For critical applications I imagine you'd use a RAID of flash disks just like a RAID of magnetic drives.
      Yeah, I wonder... stick them all in a Mirror RAID and you'll be writing to each of them at the same rate, using up their rewrite cycles simultaneously. And when the Grim Writer comes, it will come for the entire array, not just one card. Granted, they won't fail at exactly the same write, but it's gonna be a close call - too close?
    12. Re:What about the limited number of writes? by TheRaven64 · · Score: 4, Informative
      For critical applications I imagine you'd use a RAID of flash disks just like a RAID of magnetic drives.

      The nice thing about Flash is that after a cell has failed, it just becomes read-only. You can get around this quite easily in the OS by just marking the failed block as bad in your inode list. Over time, your flash drive will shrink in capacity. When it gets too small, you just copy it over to a new one and repeat the process.

      --
      I am TheRaven on Soylent News
    13. Re:What about the limited number of writes? by zippthorne · · Score: 1

      Probably. ICs that are made in the same batch are often far more similar (like, several orders of magnitude) than ICs that are made months apart. I expect you'd be ok if you started the replacement cycle early based on your best guess to stagger the storage devices and just tossed the partially "used" drives if reliability was critical.

      --
      Can you be Even More Awesome?!
    14. Re:What about the limited number of writes? by Anonymous Coward · · Score: 0

      Um, why not just buy extra devices and rotate them in and out of the lineup at various intervals?

    15. Re:What about the limited number of writes? by JulesLt · · Score: 1

      I think your mention of media hits the nail on the head - what we do expands to fit available capacity - give me enough diskspace and my CDs will all be re-encoded in a lossless format (I mean that's such an obvious upgrade for HDD based MP3 players to offer).

      What I can see is hybrid devices, with flash memory providing a large buffer for the HD - certainly enough to have the entire OS in there, which would get us some way back to the 'instant on' days of 8 and 16 bit machines when the OS was on a ROM. My hunch is the sweet spot for that could be round about 8G (enough for the OS and all major apps).

      Could maybe even have a socket for a more 'disposable' 1G chip which would take the brunt of, say, writing out the swap space on hibernation, while keeping the expensive ones for more static content. Add in some sensible heuristics as to which user files are kept in cache and which are read direct from disk (only cache if it's been read 3 times? Don't cache media files??).

      --
      'Capitalists of the world, unite! Oh ... you have' (League Against Tedium)
    16. Re:What about the limited number of writes? by jozeph78 · · Score: 1
      For critical applications I imagine you'd use a RAID of flash disks just like a RAID of magnetic drives.

      RAID... Redundant Array of Inexpensive(?) Disks.

      This should clear up once and for all if the I stands for Independent or Inexpensive. Just add marketing!!!

      --
      Ever done a `man` on `top` ?
    17. Re:What about the limited number of writes? by Tweekster · · Score: 1

      True, but honestly thsoe files could be stored on the most slow ass medium currently in use today. hell a 1000rpm ata33 drive would satisfy those requirements. as long as it is BIG... system drives basically need to be small and fast cause the chances of you running a real system that is 32 gigs (outside of servers, i mean normal computer systems) is pretty minimal. sure 10-12 people in the world might, but for everyone else, speed is a bigger issue. get a drive that is fast and reliable (and cheap because of the size it provides) and it can make a hell of a lot of sense for a thinnish client.

      --
      The phrase "more better" is acceptable English. suck it grammar Nazis
    18. Re:What about the limited number of writes? by Khashishi · · Score: 1
      You can get around this quite easily in the OS by just marking the failed block as bad in your inode list.

      What if your inode list goes bad?

    19. Re:What about the limited number of writes? by TheRaven64 · · Score: 1

      Then the node pointing to the node that goes bad is modified to point somewhere else, and the data is copied. Since the root inode is one of the the least frequently modified blocks on the disk, you can do this a lot of times. By the time your inode list starts to become immutable, you are going to have so little writeable disk space left that you are going to want to throw the flash disk away.

      --
      I am TheRaven on Soylent News
    20. Re:What about the limited number of writes? by pioneerX · · Score: 1

      You also need to so something about lsass.exe. I don't know what it's up to but it continuously performs a few I/O writes per second.

    21. Re:What about the limited number of writes? by JulesLt · · Score: 1

      I was thinking much the same after I posted - also after thinking about it, 'thinner, smaller, longer battery life' will probably win in the laptop stakes, with larger media storage being external (if you can connect wirelessly with a 60G MP3 player, why duplicate the files on your laptop?).

      --
      'Capitalists of the world, unite! Oh ... you have' (League Against Tedium)
    22. Re:What about the limited number of writes? by fnj · · Score: 1

      Hard drives also have the same limitation (You can only change the [magnetic pole] of a bit on the hard drive a certain amout of times).

      Do you have a reference for this?

      Its just you will never reach it before the mechanics of the drive fail.

      If it doesn't limit anybody, then by definition it's not a limitation.

  8. Read yes, what about write? by MirgNave · · Score: 0, Redundant

    I can definitely see the advantages of moving to solid state. But if we're talking about using flash, don't we have write speed and quantity (limited amount of writes) to worry about?

    1. Re:Read yes, what about write? by Marxist+Hacker+42 · · Score: 2, Informative

      RTFA- their write speed is reasonable (at about half that of current hard drives, supposedly, though see below for questions about this) and on a 32GB drive with a reasonable usage pattern- well, how often do you reformat an entire drive? With over a million writes on modern flash memory, it's going to take you a while to use up all the writes this drive has.

      And now for that questionable bit, from the article: While the SSD's capacity of 32 GB cannot compete with traditional hard drives that currently offers up to 80 GB space,

      I don't know abut you, but I've seen hard drives in this price range offering up to 500GB and one USB/Ethernet external that offers 1TB at less than 2x the price. Which throws the write speed into question- if 80GB drives are considered their max.

      --
      SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
    2. Re:Read yes, what about write? by max99ted · · Score: 1
      From Imation's website


      Flash memory has a write endurance limit. This limit is the number of times the flash memory cell can be written until it can not be restored to its initial condition. The industry refers to this as the erase cycles. The endurance is rated between 10,000 and 100,000 erase cycles for different types flash memories.

      --

      Please stop APK.. you're only hurting yourself.

    3. Re:Read yes, what about write? by poopdeville · · Score: 1

      You've seen 500 GB drives in the 1.8 inch form factor for the same price? Wow, we should talk. I've been looking to upgrade my PowerBook.

      --
      After all, I am strangely colored.
    4. Re:Read yes, what about write? by Marxist+Hacker+42 · · Score: 1

      Ah, that's what I missed- 1.8 inch form factor. Thank you. You're quite correct- I've seen it in 3.5" form factor, and drives up to 160GB in the 2.5" form factor that used to be the standard for laptops.

      --
      SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
    5. Re:Read yes, what about write? by timeOday · · Score: 1
      I don't know abut you, but I've seen hard drives in this price range offering up to 500GB
      Not in a laptop form factor. 80 GB was the max on laptops for several years. Just recently they bumped up to 120. They certainly are lagging desktop drives.
    6. Re:Read yes, what about write? by demonbug · · Score: 1

      They were referring to drives with the 1.8" form factor of the Samsung flash drive. The largest drive in this form factor I found in a quickie goole search was actually 60 GB, so 80 GB is probably reasonable. I'm not sure what 1.8" drives are used in - most notebook drives are 2.5" (and are significantly cheaper and offer much larger capacities).

    7. Re:Read yes, what about write? by Surt · · Score: 1

      http://memory-chips.globalspec.com/LearnMore/Semic onductors/Memory_Chips/FLASH_Memory_Chips

      Slightly out of date (claims 100k writes, while current generation flash gets around 300k, I think, but I couldn't find a cite for that quickly).

      That should be more than adequate for a typical 2-3 year laptop lifespan.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    8. Re:Read yes, what about write? by Marxist+Hacker+42 · · Score: 1

      I own a 2GB 1.8 inch drive- but I didn't catch the idea that they were comparing form factors. 1.8 inch drives are usually used in smaller-than-notebook devices- PDAs, MP3 players, and the like. It's also known as the Compact Flash form factor, because the first drives of this size were all Solid State (and the majority still are- though 32GB is a HUGE leap forward in this respect. I use my 2GB drive to store TV shows to watch on my train commute- I can fit about 10-12 hours worth at a low resolution, or 5 hours worth at the highest resolution my PDA will display- 320x240. In 32GB, I could fit around 80 hours worth.....about a week's worth of waking hours....

      --
      SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
    9. Re:Read yes, what about write? by Marxist+Hacker+42 · · Score: 1

      Further than I thought- and I missed that this phrase was a comparison to laptop drives, as opposed to "traditional hard drives", which would be more like a 3.5" form factor.

      --
      SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
    10. Re:Read yes, what about write? by geekoid · · Score: 1

      how many times does wondiws write to the disk a day during normal use?
      5,000 time wouldn't surprise me.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    11. Re:Read yes, what about write? by Marxist+Hacker+42 · · Score: 1

      Depends on the version of Windows- but remember, that's a million writes PER BYTE. Turn off the virtual memory and most of what Windows does is reads. Also, even if you have the virtual memory on, a good flash chip controller will fragment the file (remember, fragmentation means NOTHING in terms of read/write speed on a SSD) to spread it across the disk- which means instead of 5000 writes, you get maybe one or two writes per byte.

      --
      SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
    12. Re:Read yes, what about write? by Kjella · · Score: 1

      Well, I checked now and Windows (SYSTEM et al) has done about 300MB of writes in the last 14 hours. That's 300/32000 = ~1% of the space, which means it'll have written to the whole disk once every 1400 hours or once every two months. With a conservative 10000 writes lifespan that should last 1666 years. Even if I take all my apps, I'd be less than an order of magnitude away so 150 years+. Somehow I think this is a non-issue...

      --
      Live today, because you never know what tomorrow brings
  9. In my experience... by SheeEttin · · Score: 3, Insightful

    and the promise of 95% less power use

    In my experience, promised things usually fall flat on their face. Microsoft springs immediately to mind.

    And hopefully, Flash drives will replace the current magnetic platter ones. It's kind of odd for one of the most important devices in a computer to be the only moving one (And therefore the most susceptible to damage, especially in laptops).

    1. Re:In my experience... by snilloc · · Score: 1

      95% less power use means that the product engineers will just cram more volt-sucking crap in the laptop - Battery life will not be enhanced due to feature-creep.

    2. Re:In my experience... by timeOday · · Score: 1
      Let's say it consumes twice the power they estimate... then it will "only" be a 90% reduction in power consumption.

      Since a flash drive will consume no power while idle (without a 2-3 second spinup hit like a suspended disk), I can believe it.

    3. Re:In my experience... by CRC'99 · · Score: 1

      It's kind of odd for one of the most important devices in a computer to be the only moving one (And therefore the most susceptible to damage, especially in laptops).

      Of course the other most important device that still moves in your laptop will be the fan to stop those energy efficient CPUs from melting out of the case, through the table and burning your leg.

      --
      Sendmail is like emacs: A nice operating system, but missing an editor and a MTA.
    4. Re:In my experience... by Anonymous Coward · · Score: 0

      In my experience slashdot users are usually talking out of their ass. You spring immediately to mind.

      Hopefully slashdot will replace it's users with monkeys, who are better suited to this kind of idiocy. It's kind of odd that the most important part of slashdot comments is left to dimwitted users who lack the mental capacity to tie their own shoelaces, let alone operate a computer.

    5. Re:In my experience... by SheeEttin · · Score: 1

      You spring immediately to mind

      Thank you.

  10. I'd buy it by Eightyford · · Score: 4, Insightful

    I'd buy it. All that is needed is a wireless link to a network attatched file server. 32 GB holds a lot of non-multimedia files.

    1. Re:I'd buy it by symbolset · · Score: 1
      Immune to vibration and high-G stress, low power. Perfect media for carcompy and a UAV.

      I'll take two when they get to be affordable. 16Gb flash chips are selling for $41, this thing would use 16 of them. That's $656 just for the minimum amount of NAND. I think I'll wait 'till Christmas.

      Is there a SATA version? IDE is so last millenium.

      --
      Help stamp out iliturcy.
  11. Let me know when they come up with by XxtraLarGe · · Score: 1

    a 128 GB at an affordable price. I'm already near capacity on my 80 GB HD as it is...

    --
    Taking guns away from the 99% gives the 1% 100% of the power.
    1. Re:Let me know when they come up with by Anonymous Coward · · Score: 0

      if you burned off some of your porn onto DVDs, you wouldn't be at capacity nearly so often. You could label the CDs things that people would never look at too, like "Celine Deon tour photos" or "Save games". Just tryin to help out.

    2. Re:Let me know when they come up with by xerid · · Score: 1

      What is the physical size of your HD? If you're talking about a 3.5, you can fit a few 32GB flash drives in that space.

    3. Re:Let me know when they come up with by Anonymous Coward · · Score: 0

      But then it's not all available for immediate access... which, as I'm sure we all well know, is crucial.

  12. Warning... by Anonymous Coward · · Score: 5, Funny

    These flash drives still have very low rotational speeds. I'd wait a few years until they get them spinning a little faster.

    1. Re:Warning... by Anonymous Coward · · Score: 0

      I'd wait an additional few years so they can bring quantum computing into the equation. Imagine the benefits if they spin in two directions, at the same time.

    2. Re:Warning... by Ced_Ex · · Score: 0, Redundant

      These flash drives still have very low rotational speeds. I'd wait a few years until they get them spinning a little faster

      Very low? The rotational speed is 0 rpms!

      So, even if they got it to spin by a fraction, you will have INFINITE speed! I'm guessing they'll have it by next year.

      --
      Live forever, or die trying.
    3. Re:Warning... by Anonymous Coward · · Score: 2, Funny

      Actually, they are working on that: the rotational speed on these drives is up to twice that of older flash drives, using established technology. Going forward, we can expect additional improvements according to Moore's law: every 18 months, the speed can be expected to double.

    4. Re:Warning... by Anonymous Coward · · Score: 0

      Redundant??? Poor guy... the mods don't know subtle.

    5. Re:Warning... by Anonymous Coward · · Score: 0
      I think they should just take out the flash. On my digital camera, the flash consumes the most power by far, and what's the point of having a flash *inside* a laptop case anyways?

      Dude, those engineers are morons.

  13. Re:ouch by GoodOmens · · Score: 4, Informative

    If you RTFA you would see the target price is $750 and $1000 ... $6400 is the price of current flash hard drives in that size range.

  14. Old news by VlartBlart · · Score: 2, Funny

    I had one of these years ago.

    Ohhh, G i g a b y t e s - thougt it said megaby...

    1. Re:Old news by Ced_Ex · · Score: 1

      Booooooooooo!!!!!!!! Get off the stage!

      --
      Live forever, or die trying.
    2. Re:Old news by Maxo-Texas · · Score: 1

      Come on Alias... let's turn that frowny face upside down!

      I salute the parent post's attempt at humor.
      The world has few enough smiles.

      --
      She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
    3. Re:Old news by Anonymous Coward · · Score: 0

      Stolen any credit cards lately, Greg?

  15. Current drives only up to 80GB? by oculuses · · Score: 0, Redundant

    From TFA, "While the SSD's capacity of 32 GB cannot compete with traditional hard drives that currently offers up to 80 GB space..." So if the author does not know how big current drives Really are, how accurate could we expect the other numbers to be?

    1. Re:Current drives only up to 80GB? by Anonymous Coward · · Score: 0

      Author was referring to small form factor laptop drives

    2. Re:Current drives only up to 80GB? by MirgNave · · Score: 2, Informative

      I believe he was probably talking about the upper range of commonly available laptop form-factor drives.

    3. Re:Current drives only up to 80GB? by joeygb · · Score: 5, Informative

      The author is talking about 1.8" hard drives like what is used in the iPod. I don't know about you but I have seen Apple selling any 400gb iPods yet...

    4. Re:Current drives only up to 80GB? by Godji · · Score: 1

      While many of us have drives > 200 gb in our desktops, laptop drives are still (on average) around 60 / 80.

    5. Re:Current drives only up to 80GB? by joeygb · · Score: 1

      whoops.... s/have/haven't

    6. Re:Current drives only up to 80GB? by Anonymous Coward · · Score: 0

      syntax error: unterminated s/// :-)

    7. Re:Current drives only up to 80GB? by cylcyl · · Score: 2

      >> I don't know about you but I haven't seen Apple selling any 400gb iPods yet...

      I know I'm being anal, but while Apple hasn't been selling 400gb iPods, they do sell 480 Gb (60 GB) iPods :)

  16. Um Guys? by Anonymous Coward · · Score: 0, Offtopic

    http://www.simpletech.com/oem/zeus/index.php

    256GB, SATA, 60MB/Sec, 7 year warrantee.

    this is way old news, and while it was amusing to see this "article" on digg.com, it's just plain sad to see it here.

    1. Re:Um Guys? by drinkypoo · · Score: 4, Informative

      The device you link to has only 2.5" and 3.5" form factors available. This device fits in a 1.8" form factor. Nice try, though. I can see why you post as an AC.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    2. Re:Um Guys? by Anonymous Coward · · Score: 0

      Those are 2.5" and 3.5" drives. The one in TFA is 1.5". Just noting the difference.

  17. I know this is going to be asked-- by Anonymous Coward · · Score: 0

    Flash drives have come a long way, and can stand up to millions of rewrites now (if not indefinite.) The lifespan and reliability is just as good, if not better than standard hard drives.

    I'm also excited about the possibility of flash drive-based camcorders so we can get rid of DV tapes altogether... And 32 gigs is just about the point where that's a possibility (hard drive camcorders exist today, but so far with less than stellar results... One problem being a tendency to hear drive chatter on the audio.)

  18. Re:What about burnout? by OctoberSky · · Score: 1

    Whats "Multiple Times"?!?!!?

    Like 600? Or 600,000,000? If it were the later I don't think there is any reason for your post.

  19. A step in the right direction... by Zemrec · · Score: 1

    ...but still too expensive for the common user. $750-1000 for 32 GB...I think not. And while the performance numbers they quoted are better than the average laptop sized HDD and about on-par with desktop dized ones, 57/32 MB/s still isn't even saturating the bandwidth of modern hard disk interfaces.

    When are we going to see flash type drives that are cheap AND super fast? After all, secondary storage is perhaps the only remaining perfomance bottleneck in computers these days (well, that and crappy ISPs that don't/can't give you more than a few mips up and a little more down, but I digress.)

    1. Re:A step in the right direction... by RevMike · · Score: 1
      When are we going to see flash type drives that are cheap AND super fast? After all, secondary storage is perhaps the only remaining perfomance bottleneck in computers these days (well, that and crappy ISPs that don't/can't give you more than a few mips up and a little more down, but I digress.)

      If secondary storage is that fast, how is different than primary storage? The real interesting thing is that as persistant ram improves the distinction between primary and secondary will gradually fade. Or, perhaps, both will be lumped together and compared to on-die cache.

    2. Re:A step in the right direction... by cnettel · · Score: 1

      I would bet that the access time (basically none compared to a spinning disk) would help many common applications where disk access matteres, like application start.

    3. Re:A step in the right direction... by TheRaven64 · · Score: 1
      About a decade ago, I had a palmtop computer. I bought a 128KB flash card for it that cost £30. Now, I am about to buy a 1GB card for about the same price. Oh, and the new card is a fraction of the size. In 12(?) years, the amount of flash you can buy for £30 has doubled 12 times. Next year, I fully expect to be able to get 2-4GB for this much - enough for a handheld computer.

      Of course, the trend can be looked at the other way too. The price halves every year. So, if 32GB costs $750 now, it will cost $375 next year, $188 the year after, $94 the year after, and then it's in the price range of most computer owners.

      Oh, and I'm simplifying a lot here, because a lot of price reductions / capacity increases happened at the end of this 12-year period. Once flash-based MP3 players and USB drives became popular, flash became a much larger industry and R&D funding poured in. With people like Apple pre-ordering companies entire production output of flash chips, this seems likely to continue.

      --
      I am TheRaven on Soylent News
    4. Re:A step in the right direction... by el_womble · · Score: 1

      Here is what I think my next MacBooks storage specs will look like:

      4GB RAM (1GB RAM Disk for compiles and caching)
      32GB SSD (OS / Apps / Work)
      160GB External HDD (iPod for media and backup)

      My current PowerBook only has 40GB of storage, that's already more than I need, as I have a 10GB Ubuntu partition and 7GB spare. I don't need more than 20GB of persistant storage because all I do on it is web/email/code/office. All my photos, music, movies etc. are stored on my iMac. If I'm at home I use AFP to connect to my media, if I'm on the move I use my iPod. The only thing that stops me from recommend this situation to everyone is the DRM in iTunes.

      I can't be that unusual. I have an iMac at home, PowerBook everywhere and a Dell at work. The only thing that makes me even remotely 'on the edge' is the use of a laptop, the majority of computer users I know have a t least a desktop at home and at work. Why can't I register my iMac as a central repository and be able to elect 5 other computers as subrepositories (ie, I can freely move music from my G5 to the PowerBook or Dell using either the iPod or the magic of the intar-web, even if they are registered with iTMS to play the music?). By the same token, why can't I rip a CD on the PowerBook, and have it magically appear on my iMac when I resync the iPod?

      --
      Scared of flying, pointy things snce 1979!
  20. Not the biggest power eater by Dwedit · · Score: 5, Informative

    During heavy disk read activity, the HD is only uses 15% of all the power. (source) The real key to decreasing laptop power consumption is dimming the screen, which can reduce power consumption percentage from 26% down to 7%.

    1. Re:Not the biggest power eater by Zaffle · · Score: 1
      During heavy disk read activity, the HD is only uses 15% of all the power. (source) The real key to decreasing laptop power consumption is dimming the screen, which can reduce power consumption percentage from 26% down to 7%.

      Dimming the screen is something that the user can do, and can be done easily. Reducing disk power usuage isn't as simple as turning the brightness knob down (damn how I miss those knobs. Nothing has knobs now). If the HD power usage is dropped from 15% of total consumption at peak load to 1%, then thats a huge power saving.

      And besides, it doesn't matter what the heavy usage rate is, its the idle that counts. My laptop would be under heavy usage about 10% of the time. The other 90% its idling or doing little activity (typing, mouse movements, etc). An idle spinning hard disk still uses heaps of power, and spinning down the disk means it takes time before it speeds up.

      I would love to see a "hard disk" that is actually a non-volitile storage system (eg HDD, flash, etc), fronted with battery backed up RAM. The battery will provide power even after shut off, and will provide enough power to write the entire contents to non-volitile storage. This is why flash would work best, it takes much less power to write to flash than it does to spin a HDD and write to it.

      A Flash/RAM/Battery smart storage system would have an effective read/write speed limited only by the BUS that it was connected with (under the assumption that a prepherial bus will never be as fast as RAM).

      The only reason its not done yet is:

      • Too much power is required to power a HDD on battery effectivly and long enough to write the whole disk
      • RAM prices haven't matched HDD prices (16GB of RAM isn't cheap)
      • Flash drives weren't big enough. Though 16GB is enough for a boot/app system, writing user data to convential HDD.

      Hmmm. I should patent that idea. Better yet, this comment can be considered prior-art against such a patent if I don't beat you bastards down to the patent office.

      --

      I use to have a funny sig, but slash cut it off, and I forgot what the punchline was.
    2. Re:Not the biggest power eater by bogomipz · · Score: 1

      You got the HD and LCD percentages from different charts. When the HD consumes 15% the LCD consumes 19% (when the disk spins up, the percentages for the other parts go down), which is not so far apart. If you turned down the LCD so it only used 7% (actually this would be lower as well, still because of the increased HD consumption), the HD would draw approximately 20%. 19% -> 5% for the LCD is a good thing, and so is 15% -> 2% for the HD. I pulled 2 and 5 out of the air there, though. Let's just say that all the parts that draw about 10% (LCD, HDD, CPU, GPU, wireless, etc) are worth making more efficient.

  21. Technology currently in use already by Furp · · Score: 4, Informative

    This technology has already been put to use in a commercial environment, and has given outstanding performance from what I've seen. The game EVE Online http://www.eve-online.com/ has already done this with their clustered servers and greatly reduced the lag. Keep in mind that this is a game where there is only a single universe (No shards or other servers) and they quite often push over 20,000 simultaneously logged in accounts at a time.

    When placed in the right environment, this technology just screams. A good example would be for huge database operations that have hundreds if not thousands of concurrent accesses. The databases that maintain the pay information for the US Military come to mind easily.

    1. Re:Technology currently in use already by vidarh · · Score: 5, Informative
      I very much doubt they use flash based SSD's for performance enhancements.

      Most large scale systems that use SSD's to increase DB performance do so using DRAM (mainly) or SRAM based units with battery backup, RAM based RAID and controllers that dump the data to disk either on an ongoing basis or in the case of a power failure (using battery power to keep things up at least long enough to write a consistent snapshot to disk).

      The units are ridiculously expensive, but far faster than anything you'd manage to get with flash or harddisks (typically they're maxing out the controller/bus you connect to them via).

    2. Re:Technology currently in use already by kindbud · · Score: 1

      We were using battery-backed RAM-based disk drives for database indexes 10 years ago. The news here is that flash RAM is finally catching up in terms of cost and performance, not that memory-based disks speed up database performance. That is old news.

      EVE servers are using DRAM-based storage, not flash-based. They specifically mentioned the product they are using, too, when crowing about the recent upgrade: http://www.superssd.com/products/ramsan-400/

      --
      Edith Keeler Must Die
    3. Re:Technology currently in use already by Furp · · Score: 1

      I had found that as well, as when I wrote my reply I was operating on memory, as there were a number of filters in place. And after doing some research after, I found that I was wrong.

      Eve-Online is running off of RamSan-400. RamSan-400 is a RAM based storage device with battery backup. Google provides easy links to information on it. The only part that I regret is this: I had to dredge it up using Lynx through the eve-online.com site.

    4. Re:Technology currently in use already by pthisis · · Score: 1

      Most large scale systems that use SSD's to increase DB performance do so using DRAM (mainly) or SRAM based units with battery backup...The units are ridiculously expensive, but far faster than anything you'd manage to get with flash or harddisks

      You can get versions in the $150 range that use standard SIMMs, then fill them up with whatever size you want. Definitely not the same as the high-end stuff, but pretty cool. Gigabyte's i-ram comes to mind--it's battery-backed, kind of (like 12 hours completely off power or something, draws from the PCI bus to maintain state/recharge).

      http://techreport.com/reviews/2006q1/gigabyte-iram /index.x?pg=1

      --
      rage, rage against the dying of the light
    5. Re:Technology currently in use already by Anonymous Coward · · Score: 0

      Like the RAMSAN - http://www.superssd.com/

    6. Re:Technology currently in use already by Shanep · · Score: 1

      Most large scale systems that use SSD's to increase DB performance do so using DRAM (mainly) or SRAM based units with battery backup, RAM based RAID and controllers that dump the data to disk either on an ongoing basis or in the case of a power failure (using battery power to keep things up at least long enough to write a consistent snapshot to disk).

      Yes, in 1996 the stock exchange I was working for started using a DEC Storageworks product, which was essentially a 1GB RAM disk with 1GB hard disk mirroring the RAM for backup, used in a great big VAX that's bigger than my fridge. I think it was 1GB, it might have been 2GB. Either way, it seemed pretty awesome at the time. I'd trust that for high transaction tasks much more than flash based storage at the moment.

      --
      War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
    7. Re:Technology currently in use already by vidarh · · Score: 1
      These days you can get up to about 1TB in a standard 42U rack... But given that I was quoted up to $250k (depending on options) for a 32GB unit, I don't even want to think about what they charge for the 1TB unit...

      Of course, there are reasons why these types of units are so expensive, like the fact that most of them use hardware RAID with chipkill support (the DIMM's will automatically be taken out of use if they get unreliable, and can be hotswapped), two or three separate power supplies, typically multiple fibrechannel interfaces that can be max'ed out at the same time for interfacing with the box (or in some cases custom PCI cards with custom buses).

  22. Re:What about burnout? by Soporific · · Score: 2, Informative

    Someone had posted this on another flash drive story here but it basically went that if you reserved 10% or so of the drive simply to keep rotating blocks it would last as long as a hard disk, more or less.

    ~S

  23. Re:What about burnout? by Anonymous Coward · · Score: 0, Redundant

    They do, but modern flash has something in the range of ~100000 writes per sector. Magnetic hard disks get about 10 times as many before they fail, but given the average hard disk is running a Windows OS which needs defragging the difference isn't all that big.

  24. Nice estimated price... by ShyGuy91284 · · Score: 4, Informative

    It estimated to cost$700 - $1000. While this may seem like a lot, for something new, this isn't. I remember reading how much a hardrive would have cost for an old IIGS that had maybe 8 disks worth of storage space I think. And although expensive, $700 isn't expensive enough to be out of the reach for consumers. Just expensive enough to be out of the reach for most sane typical consumers.

    --
    In undeveloped countries, the consumer controls the market. In capitalist America, the market controls you.
    1. Re:Nice estimated price... by pthisis · · Score: 1

      I've been waiting a long time to get a nearly silent drive for my home machine. I'm willing to pay a bit extra for that feature.

      This seems at about the crossover point from "flash devices are only for expensive specialty use by corporations/research projects/etc" to "early adopters may now be willing to shell out for it". Presumably once they get the ball rolling, production ramps up and prices fall.

      --
      rage, rage against the dying of the light
    2. Re:Nice estimated price... by Hadlock · · Score: 1

      Makes sense. Anyone who's taken a marketing or buisness class (or hell, just follows technology trends/prices) understands that when you have a hot new technology, you charge serious $$$ for it, trying to recoup your R&D as much as possible, and to bleed the early adopters dry, because, well, you charge what the market will bear. Ultimately, A/V people will buy these. Buisnesses will buy them for their purposes, and you'll start to see them in super high end laptops as the price begins to drop over the next two years. After three to five years, with the initial costs (manufacturing, R&D) paid off, the price will drop so that they can offer their product to a different market, ala the consumer market.
       
      Does this happen often? Heck, I probably could have saved myself some time and just copied and pasted a similar argument from the early-flat panel LCD era on slashdot. Or the 4GB compact flash era. Or...
       
      Flash drives are the Next Big Thing. In three years' time, Western Digital and Seagate will be dropping their prices to stay competitive with Samsung's superior technology. In six years you'll be able to tell middle schoolers "i remember when ALL computers came with hard drives that spun on big (3.5) magnetic discs!". Floppies died two years ago, hard drives will bit it shortly.

      --
      moox. for a new generation.
  25. RTFA! by RingDev · · Score: 4, Informative

    Price point on the 32gb drive is expected to be $750-$1000. The $6400 product is a currently available military grade drive. It'll take a wee bit more abuse and temperature range then the 'cheapest bidder' built one that will hit the commecial market.

    -Rick

    --
    "Most people in the U.S. wouldn't know they live in a tyrannical state if it walked up and grabbed their junk." - MyFirs
    1. Re:RTFA! by Kjella · · Score: 1

      The $6400 product is a currently available military grade drive. It'll take a wee bit more abuse and temperature range then the 'cheapest bidder' built one that will hit the commecial market.

      Not to mention the most important reason. Because you can.

      --
      Live today, because you never know what tomorrow brings
    2. Re:RTFA! by Lord+Kano · · Score: 0

      The $6400 product is a currently available military grade drive. It'll take a wee bit more abuse and temperature range then the 'cheapest bidder' built one that will hit the commecial market.

      They're pretty much the same things. Military contracts tend to go to the lowest bider.

      LK

      --
      "Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
  26. With compression you might get 50 gig by ScrewTivo · · Score: 2, Interesting

    and with the speed increase not see a difference. I have wanted this since my first 286.

    1. Re:With compression you might get 50 gig by creepynut · · Score: 1

      Except the speed increase is in disk read/possibly writes. Compression could, and probably would weigh heavily on the CPU, creating a brand spanking new bottleneck.

  27. This is big news. by zymano · · Score: 3, Insightful

    Only price is the barrier now for the slllloooooooowest parts of a computer.

    1. Re:This is big news. by Surt · · Score: 1

      Unfortunately, it's only twice as fast. :-(
      But maybe it will get faster yet.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    2. Re:This is big news. by ahaveland · · Score: 1

      Unfortunately the operator will probably always be the slowest part of the computer, unless extensive genetic modification becomes possible.

      With an AMD 4200+, at least my spare cycles are BOINCing on something useful while I think about the next lines of code or whatever else I do...

      Yes, I would love a Tb of Flash mem, but the problem of organising and storing backups will still grow to match developments.

  28. Re:What about burnout? by jnd3 · · Score: 4, Informative

    Most flash can handle something like 100,000 erase cycles. And most flash file systems have wear-leveling algorithms to ensure you're not hitting the same sectors over and over. Even with standard usage they should be good for several years at the very least.

  29. a little outdated? by moochfish · · Score: 0, Flamebait

    the SSD's capacity of 32 GB cannot compete with traditional hard drives that currently offers up to 80 GB space

    In the other news, this journalist can't compete with modern journalists who have been known to use "Computers."

  30. Man the RPM of these SSDäs are slower than my by mOOzilla · · Score: 1

    These wont burn the palm of me with my 7200 RPM 100GB drive I have in here at present. Problem I worry about is reweite lifetime. We wont have to defrag would we as the read time from any location is constant from any memory address, correct? Its a great step however I can see HYBRID drives being big news rather than 100% SSD except on "web tablets comebacks" and Viao type devices. For the power munchers here with Ferrari's Hybrids would be welcome.

  31. WTF? by caffeination · · Score: 1

    They ought to put the number 14967401 in the dictionary definition of 'tanking'. I certainly hope you won't be here all week :-)

  32. Re:ouch by Alex+P+Keaton+in+da · · Score: 1

    Ouch- keep in mind that my 6 megapixel camera, if available a few years ago, would have cost more than some cars....
    Volume production will bring prices down. Remember the $2000 component CD burners in the late 90s?
    I can't wait until I can afford all these GBs of flash tastiness!!!!

    --
    And All I Ask is a Tall Ship And a Star to Steer Her By
  33. Re:ouch by Anonymous Coward · · Score: 0

    Actually, no.

    If you RTFA you would see that $750-$1000 isn't the target price, it's an educated guess by the article's author.

  34. NO NO NO NO NO! by chrysrobyn · · Score: 4, Funny

    I'm trying to close on buying a house! And Samsung, Apple's iPod Nano flash supplier comes out with this?

    APPLE, please PLEASE do not come out with an Intel Mac portable featuring a flash drive (with its tasty power consumption, lower power and low low low seek times) after I clean out my savings! I would have been exceptionally happy to have a PowerPC flash computer a year ago or 6 months ago, or even maybe 3 months ago, but I'm cleaning out my savings here for the part of a house that the bank won't cover!

    Wait 6-12 months for a flash based portable and I'll forgive you for going to Intel.

    1. Re:NO NO NO NO NO! by geekoid · · Score: 1

      yeah, like in 6 to 12 months you'll be able to afford it.
      Your surplus cash days are over the same time your escrow begins.

      Don't worry, it's worth it.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    2. Re:NO NO NO NO NO! by chrysrobyn · · Score: 1
      yeah, like in 6 to 12 months you'll be able to afford it. Your surplus cash days are over the same time your escrow begins. Don't worry, it's worth it.

      I'm not sure you did it right. When I bought my first house, we found one we could afford to make payments on and still put something away at the end of the month for emergencies. When enough emergencies don't happen for long enough, we buy something like a TV. Since the end of 2003 and the first half of 2004 had enough emergencies that we got in the habit of just saving all we could, we now have enough to be able to get a slightly bigger house far closer to my work.

      If your "surplus cash days are over", then you can't afford to lose your job. You can't afford to have any car problems. You can't afford a new sofa, TV or kitchen table (whatever you don't have) to go in your shiny new house. For me, my surplus cash days started the day after I closed on my first house. A responsible mortgage payment equals long term stability and security, even if it isn't a McMansion in the 'burbs.

  35. That guy would buy it. by Anonymous Coward · · Score: 0

    That guy would but it. Seriously.

  36. Re:ouch by jandrese · · Score: 1

    $1000 == Ouch. Definatly a high end laptop feature. That's more than many people pay for their whole laptop.

    --

    I read the internet for the articles.
  37. Amazing... by butterwise · · Score: 0

    32GB? That's a LOT of pr0n...

    --
    If a baby duck is a "duckling," why would anyone want to eat "dumplings?"
    1. Re:Amazing... by Anonymous Coward · · Score: 0

      32GB? That's a LOT of pr0n...

      Amature.

  38. Star Trek by orty78 · · Score: 5, Funny

    I'm reminded of Star Trek. We all know that Star Trek is the way of the future. Talk about beating a dead horse. But this story made me think back on the episode where Cmdr. Data is swapping all of those USB flash drives into a different order to overcome some technical problem. USB and Flash memory are therefor, conclusively, here to stay for good.

    1. Re:Star Trek by Anonymous Coward · · Score: 0

      Ummm... those were isolinear chips (holographic data storage), not flash storage. And Data wasn't putting them in a new order to "overcome some technical problem"... he was putting them all back into the command computer so that the egines could be re-activated and they could escape from impending doom. (This is because another crew member, afflicted with a virus, removed all the chips in a fit of drunk-like behavior. This is from the episode The Naked Now.)

      So I think it's pretty clear that flash memory is not "hear to stay" but merely a stepping-stone towards the true "technology unchained" that Star Trek predicts...

      ....ummm.... I mean ... it's just a TV show.

    2. Re:Star Trek by Akaihiryuu · · Score: 3, Interesting

      The isolinear chips are actually an optical storage format. I guess you could kind of compare them to CD-RW, only they are a LOT faster, and they also store their data in 3 dimensions instead of 2. Also, a lot less effort has to be made in ST computers to make them use an optical format, because the computers themselves are optical, not electronic. Crystals instead of silicon chips, fiber optics instead of semiconductors and connecting wires. They also do some funky stuff like putting a warp field around the computer so that the fiber optics can exceed the speed of light. Actually, Star Trek tech optical tech is kind of similar to Ancient and Goa'uld (since the Goa'uld scavenged Ancient tech to come up with their stuff) tech in Stargate.

    3. Re:Star Trek by geekoid · · Score: 1

      "Actually, Star Trek tech optical tech is kind of similar to Ancient and Goa'uld (since the Goa'uld scavenged Ancient tech to come up with their stuff) tech in Stargate."

      true...there both fictional.
      Plus, warp does not make the ship go fater then light. It makes it so it doesn't travel all the distance. Since the bulk of the distance would be in the wiring getting to the display, it makes no sense to put a warp field around it.

      Sheesh.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    4. Re:Star Trek by iibagod · · Score: 5, Funny

      See, this is why you NEVER, EVER, post anything remotely funny about Star Trek. Let's see....3 comments pointing out that you're wrong, dead wrong, those were ISOLINEAR chips, dumbass....and 1 comment about how the robot gets laid.
      Something about ST just seems to bring out the worst part of geekdom.



      Now all the TREKKIES are going to mod me down because I called Data a ROBOT when he's clearly an ANDROID. Karmic Suicide.



      disclaimer: I'm currently watching the entire run of ENT start to finish. But I actually get away from the screen .....for HOURS at a time, even!

    5. Re:Star Trek by Sj0 · · Score: 1

      That was nerdy even by Slashdot standards.

      --
      It's been a long time.
    6. Re:Star Trek by Sj0 · · Score: 1

      The sad part is that Star Trek engineers have obviously never heard of fail safe engineering, so running the warp drive through the same computer they use to play "Doom 274: The revenge of the return of the wrath of Cyberdemon" isn't out of the realm of possibility.

      Those who would disagree with me should remember all the times a failure made a process do something dangerous. :P

      --
      It's been a long time.
    7. Re:Star Trek by Zontar_Thing_From_Ve · · Score: 2

      Plus, warp does not make the ship go fater then light. It makes it so it doesn't travel all the distance.

      While I partly shudder at the thought of arguing about a fictional TV show, you, sir, are dead freakin' wrong. Warp drive DOES indeed go faster than light. Had you watched more than a handful of Star Trek episodes, you would clearly understand this. To educate yourself, please see:
      http://en.wikipedia.org/wiki/Warp_drive_(Star_Trek )
      You have obviously confused warp drive with some type of jump drive technology which does not involve faster than light travel but makes ships jump large distances essentially instantaneously.

  39. or, an HD that works above 12,000 feet. by sampas · · Score: 2, Interesting

    It will be nice to have the additional capacity on GPS devices and tablets used for aircraft navigation. Traditional HD's have trouble above 12,000 feet because the head's "wings" don't produce enough lift at lower pressure.

    My question is how many write operations is it rated for? Others list 300,000 -- is that a lot or a little?

    1. Re:or, an HD that works above 12,000 feet. by Echnin · · Score: 1

      Aren't the HDDs in a compressed environment in the aircraft though?

      --
      Lalala
    2. Re:or, an HD that works above 12,000 feet. by xerid · · Score: 1

      You know too much about aircrafts, you damn terrorist!

    3. Re:or, an HD that works above 12,000 feet. by 2443W · · Score: 1

      Depends on the plane, airliners are compressed but a lot of other planes aren't, especially small general aviation aircraft.

    4. Re:or, an HD that works above 12,000 feet. by Timothy+Brownawell · · Score: 1
      "Aren't the HDDs in a compressed environment in the aircraft though?"

      Yes, it's pressurized, but not to the same pressure as at ground level (that would take too much work). I sort of recall hearing that HDDs can have trouble if they live on high mountains, too.

      Tim
    5. Re:or, an HD that works above 12,000 feet. by sampas · · Score: 2, Informative

      My old Piper Cherokee isn't pressurised at all. I can get an expensive navigation system, but if I want to use a less-expensive laptop, I can't take it above aboute 10-12,000 feet without the risk of crashing the heads.

    6. Re:or, an HD that works above 12,000 feet. by wed128 · · Score: 1

      What they ought to do is give the HDs an airtight seal, so the pressure around the actual disk stays constant.

    7. Re:or, an HD that works above 12,000 feet. by Profane+MuthaFucka · · Score: 1

      OK then, rule of thumb. If the pilot offers you oxygen, don't turn on your computer. And don't smoke either.

      --
      Fascism trolls keeping me up every night. When I starts a preachin', he HITS ME WITH HIS REICH!
    8. Re:or, an HD that works above 12,000 feet. by Zebadias · · Score: 1

      How would you balance for temprature change if it is sealed?

    9. Re:or, an HD that works above 12,000 feet. by nmg196 · · Score: 1

      > What they ought to do is give the HDs an airtight seal, so the pressure
      > around the actual disk stays constant.

      Err, no, definitely not a good plan! If you change the temperature by just a couple of degrees, the pressure changes dramatically. The temperature of a hard disk goes up enormously over the first few minutes after you power it up. If you completely sealed it, the pressure in the disk would be very high after a few minutes but would be too low if you cooled it down a bit or turned it on when it was very cold (eg in an aircraft). All hard disks have a small vent hole somewhere that usually says "do not cover". When you've got a spare hard disk, try covering that hole and seeing what happens (hint: data loss will occur, to put it mildly).

    10. Re:or, an HD that works above 12,000 feet. by evangellydonut · · Score: 1

      I guess you have no clue on how the aerospace industry works and the problems associated with it... 12,000 ft may not have enough air pressure to spin up harddrives, but it certainly has enough radiation to cause tons of errors in the flash device. do you want your plane to crash due to a "bit-flip" (aka seu-event)?

    11. Re:or, an HD that works above 12,000 feet. by sampas · · Score: 1

      Actually, I'm a pilot and regularly update the navigation database in my GPS. There are plenty of flash memory devices inside aviation Global Positioning Systems, some of which are FAA-certified for IFR flight and are coupled to auto-pilot systems. There are even complete guidance systems that run on a computer. (Although all planes with those also have the backup "steam guages," and I for one, would not go totally electronic yet.) Most small-plane crashes are due to pilot error rather than mechanical failures, however.

      What the cheaper, larger-capacity hard drives will do for me is let me build my own system (similar to this) that could do additional things, like hold approach plates. Best of all, I could do it all for far less.

      But you're right, I really don't know how our aerospace industry works, I just know how to fly, navigate, and communicate using various devices. If you're that worried about radiation affecting any flash drives at altitudes, you probably shouldn't be flying on any modern airliners.

      PS: I learned to fly when I got sick of airsims and wanted to do the real thing.

  40. FlashBelt by Doc+Ruby · · Score: 1

    How about a "FlashBelt" which can take 10+ flash chips, a flat rechargeable battery, and Bluetooth for storage accessible by mobile phones and visited terminals? $100 32GB chips would be great, but even just forcing down the price of 10GB chips below $10 would make personal storage more "intimate".

    --

    --
    make install -not war

  41. Where I see SSDs in laptops being used most. by Demon-Xanth · · Score: 2, Insightful

    Ruggedized applications.

    Example: a mechanic using it to interface with a car's OBD port.
    He's not going to be writing to the HD a while lot, but you know damned well that it's not going to be treated lightly. 32GB is pleanty large to put and OS and the diagnostic/tuning apps on.

    Make that laptop low enough power to plug into a cigarette lighter and you got a nice tool.

    Another example: Some geologist needs to take data off of some geophones in the middle of places with names like "Desolation Wilderness". A laptop with a longer battery life and a HD that is going to survive being in a backpack is going to make things alot easier. Hiking out 10 miles to the middle of nowhere isn't something that you want to have to re-do because something broke or you ran out of battery life.

    I don't forsee anyone having one at the next LAN party. Though given the number of people with hilarious setups, it could happen. Afterall, who'd buy a 150GB HD that cost $350? (WD Raptor)

    --
    If you think education is expensive, you should try ignorance -- Derek Bok, president of Harvard
    1. Re:Where I see SSDs in laptops being used most. by Anonymous Coward · · Score: 0

      Er, for OBD tuning and diagnostics? 32 gigs? Man, all you need is 32 megabytes for the kernel (Linux, BSDs, whatever) and the application (say based on embedded Qt), and whatever your logging demand is. It might be actually easier to use an external USB pendrive for logging purposes, as it will be likely cheaper than onboard flash memory. There are some nice embedded USB hosts around, in case your embedded chipset/platform doesn't come with USB.

      I can't fathom why people think that something so rudimentary and stand-alone as OBD logging/diagnostics/tuning necessitates the bloat of say XP or Vista or whatever. Even something like Fedora Core would be a very bloaty thing to put in such a device.

      You don't really need nor want anything besides what the tech needs to do the job, and that typically doesn't imply the whole darn OS install. Who needs a yet another copy of Solitaire or Tux Racer on one of these ;)

      Cheers, Kuba

  42. Re:ouch by demonbug · · Score: 1

    Which,the article adds, are considerably more robust (at least in terms of temperature ranges) than what Samsung is shooting for - which might be part of the price difference. On the other hand, if you just took 32 1 GB CF cards you would be paying around $1100, which makes their target price sound very reasonable (although far too expensive to be widely adopted I would think - I could see possibly paying double or even triple the price for the advantages, but not ten times as much for something of questionable utility for most people).

  43. Re:ouch by Sebilrazen · · Score: 3, Informative

    That's probably Taiwanese dollars, that's where the byline is.

    0.0308676(Taiwan/US) * 6400(Taiwan) = 197.55264(US)

    That's still $6.20 US/GB so still not very desirable, but if they can EoS down, and get the battery life trade off it may be worth it.

    --
    "There are no facts, only interpretations." --Friedrich Nietzsche.
  44. This is reasonably easy to cope with by Anonymous Coward · · Score: 0

    DamnSmallLinux is designed to run off compact flash cards. Most operations happen in RAM with a backup to the flash happening infrequently. http://www.damnsmalllinux.org/ I've been using it for a thin client for a few months now with no problems. It amuses me to be able to have a spare drive that's cheap and easy to pop in if I mess something up.

  45. Re:ouch by ClockN · · Score: 0

    I am thinking the $6400 question relates to; My question is how many write operations is it rated for? Others list 300,000 -- is that a lot or a little?

    --
    There are 10 types of people in the world... those who understand binary and those who don't.
  46. Not relevant... by MadCow42 · · Score: 1

    Defragging isn't relevant for a flash memory based device... you can have data spread all over the place, and it shouldn't affect your read/write speed like it does on a hard drive. The physical seek time of the head is what causes fragmentation on a hard drive to be a problem. No head, no problem.

    MadCow.

    --
    I used to have a sig, but I set it free and it never came back.
    1. Re:Not relevant... by temojen · · Score: 2, Informative

      Some filesystems (ie Reiser4) move or consolidate files (aka "defrag") in the background , and don't know what kind of block device they're on. You'd want to tell it not to do bother doing that then. Except the kernel/ATA interface still reads and writes by the block, but a block in some filesystems (Reiser4) may contain parts of several files, so you'd want to eventually consolidate files so you don't have to read/write a whole lot of blocks to access a single file which might be smaller than a block.

      A worst case scenario would be a filesystem similar to Reiser4 with consolidation turned off, and lots of files growing by small amounts frequently.

    2. Re:Not relevant... by RicRoc · · Score: 2, Insightful

      Perhaps someone could invent a file system that fits better with the new hardware. Filesystems today are designed for disc access -- tomorrow's hardware requires tomorrow's software. And I bet Reiser will be on top of that too! :-)

      --
      Who?
    3. Re:Not relevant... by Fzz · · Score: 1
      A worst case scenario would be a filesystem similar to Reiser4 with consolidation turned off, and lots of files growing by small amounts frequently.

      This is pretty easy to deal with in the SSD itself. You just need to front the drive with a large DRAM disk buffer to decouple reads and writes, and then make sure the drive firmware reads each block before writing it. There's no need to write any bytes that haven't changed. Given that reads are a lot quicker than writes this may even improve performance too.

  47. That's a feature? by pingus · · Score: 1
    "Some of the advantages offered are ... read speeds more than twice that of a normal hard drive.."

    I'm sorry, is this some kind of new feature I was unaware of?

    1. Re:That's a feature? by +InvaderSkoodge · · Score: 1

      "I'm sorry, is this some kind of new feature I was unaware of?"

      Am I missing something? Double the read speed of a normal HD sounds like a good feature to me. Systems boot faster, apps load faster, etc...

    2. Re:That's a feature? by Anonymous Coward · · Score: 0

      i suspect that was her first attempt at sarcasm

  48. Fragmentation? by Dr.+Evil · · Score: 3, Insightful

    Does fragmentation matter when there are no heads to move?

    1. Re:Fragmentation? by temojen · · Score: 1

      On Data-Journaled filesystems, yes.

    2. Re:Fragmentation? by nettdata · · Score: 1

      How so? I'd always thought that defragging allowed file segments to be (re)assigned to linearly sequential drive blocks so that they could be read much faster, as the heads have much less travelling/seeking to do.

      With this, there is no physical motion, so no apparent gains.

      Do journalled file systems have need for some sort of sequential read/write in memory addressing or physical blocks?

      --



      $0.02 (CDN)
    3. Re:Fragmentation? by kiddygrinder · · Score: 1

      Absolutely, On my NTFS drive i had one (1) ridiculously fragmented 1gb file that was taking up 10 gig of drive space

      --
      This is a joke. I am joking. Joke joke joke.
    4. Re:Fragmentation? by temojen · · Score: 3, Insightful

      Data-Journalled filesystems (eg Reiser4) keep data in the journal too, not just metadata (those are metadata-journaled filesystems eg Ext3FS), so each block may have parts of several files, and each file may be spread across many more blocks than it would fill on it's own.

      You eventually have to consolidate the data of each file. Not nescesarily to sequential blocks, but so files are not sharing blocks.

      For flash memory, non-journalled filesystems like Ext2 (mounted -o noatime) may be best. Although that still tries to keep large chunks of files sequential. It might be better to have a non-journalled filesystem that does not pre-allocate inodes and data blocks, but just keeps a free block list and allocates from it in Least-Recently-Used order.

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

      Consider X bytes of data.

      Case 1: X bytes stored in one physical block. One read required to access data.

      Case 2: X/2 bytes stored in two physical blocks. Two reads required to access data.

      Accessing flash storage is slow. I/O operations have OS and CPU overhead and device bottlenecks. Obviously you'd want to reduce the number of accesses even if there is no physical head seeking involved.

    6. Re:Fragmentation? by Shanep · · Score: 1

      Does fragmentation matter when there are no heads to move?

      The point is that the OS might THINK that a defrag is in order (OS assumes head movement latency is always an issue) and then goes right ahead and moves data around, in the process performing writes which in the case of solid state memory is not needed.

      --
      War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
    7. Re:Fragmentation? by MarkByers · · Score: 1

      The point is that the OS might THINK that a defrag is in order (OS assumes head movement latency is always an issue)

      No... users think that fragmentation is the problem, but really it's their choice of file system that is the problem. Use a modern file system. Most modern operating systems have this already.

      --
      I'll probably be modded down for this...
    8. Re:Fragmentation? by Shanep · · Score: 1

      No... users think that fragmentation is the problem, but really it's their choice of file system that is the problem. Use a modern file system.

      No... I'm not talking about users, I'm talking about some operating systems which take in upon themselves to automatically perform a defrag when it is not needed. XP can be configured to do that, newer versions of OSX (since Panther) do it by default and I believe some filesystems for Unix-like OS' can also move stuff around automatically in the name of optimization. Users are a seperate problem and if they're doing something to the detriment of their data, without any need to do that, then that is their problem. I am specificially speaking about automatic processes which assume head movement issues to make gains in optimization.

      I was not refering to users. They will always be a problem.

      Most modern operating systems have this already.

      Have what? The ability to detect the difference between solid state storage and disk/head based storage and then adjust optimization accordingly?

      --
      War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
    9. Re:Fragmentation? by Anonymous Coward · · Score: 0

      Are there really no programmers here? Of course fragmentation matters on flash drives. It's all the difference between a pointer to a block of memory and having to traverse a linked list. Fragmentation means both using more memory and taking more time to access the data, a special lose-lose scenario.

    10. Re:Fragmentation? by bogomipz · · Score: 1

      So all the fancy, new wave FS features are basically not suited for the storage devices of the future. Sounds like FAT is going to raise from its grave :p

    11. Re:Fragmentation? by silverdirk · · Score: 1

      "You eventually have to consolidate the data of each file"

      Not quite. They consolidate files now for efficiency in accessing them. If you have a device that can do random access without millisecond delays, and which is bad to have unnecessary writes, then you could just leave the files in the journal.

      In other words, the entire disk can be the journal, and all files can be contained in it.

      Imagine a file system where the entire disk is divided into stripes. You erase a stripe when its data is obsolete, and write a new stripe whenever you have enough data, combined from however many writes have occoured lately. As old blocks of data become outdated (because the block has been written again in a new stripe) you end up with stripes that are only partially filled with useful data. When you need to reclaim this space, you add all the blocks in the target stripe to the "current writes" list, so they get written again (in a new stripe). Then the entire old stripe is invalid, so you can write a new stripe over top of it. This sytem could even be used to do a round-robin on the entire drive, which would give you the longest drive life of all.

      --
      Mark of the Coder fades from you. You perform Opening on World of Warcraft. Warcraft crits GPA for 4. GPA dies.
    12. Re:Fragmentation? by wilec · · Score: 1

      Thanks for the information about the differences between Reiser4 and ext3 file systems. I had not investigated Reiser4 in detail before and did not realized there was this level of difference between it and ext3. I have choosen ext3 usually, except for a recent evaluation install of Suse10 where I choose Reiser4 just for the heck of it. I will read up more on Reiser4.

      I sure like the fast recovery time of a meta-journaled file system like ext3, I would hate to give that up. And then theres the simply security blanket thing. I have been using ext3 for several years now, with not a single non recoverable loss of anything outside of /tmp files. Of course ext2 was always pretty reliable as well but the fsck was rather tedious compared to ext3. I have had horrible luck with all the FAT derivatives, and far less that prefect luck with NTFS or HPFS. A safer file system is a large part of why I switched from OS/2 to Linux. well that and better new hardware, especially USB2.0, support on the cheap. I miss a few OS/2 apps but Linux is working out fine.

      Matthew

  49. 95%? Where? by geobeck · · Score: 1

    From TFA:

    Samsung says that the Flash disk consumes only 0.1W when not in use and just 0.5W under load. For comparison, a typical mobile hard drive consumes somewhere between 1W and 2W of power in seek, read and write processes and between 0.2W and 0.8W when idle.

    So when idle, it consumes between 50% and 85% less power. Under load it consumes between 50% and 75% less. The only 95% I see is comparing the flash drive at idle to the hard drive at full seek.

    Maybe the OP does statistical analysis for government projects...

    --
    Find environmentally and socially responsible products on http://buy-right.net
  50. 32GB and shrinking... by heretic108 · · Score: 1

    Given the 10k-100k endurance of the flash memory, even the act of booting up a machine with a main flash drive, or starting an application on it, is going to eat endurance cycles.

    Worse if you runs Windows, where you can't even scratch yourself without causing a write to the registry. You'll find your 32GB drive shrinking over the months towards zilch.

    Guess it's probably ok for portable use, where mechanical drives have a risk of sudden failure. And one can always buy a new flash drive and copy over the data when the old one gets too exhausted.

    --
    -- In the beginning was the WORD, and the WORD was UNSIGNED, and the main(){} was without form and void...
    1. Re:32GB and shrinking... by Anonymous Coward · · Score: 0

      I could see this working VERY nicely with unix style setups though. Put the os on that drive. Then symbolic link the 'log/user/data' areas to a real smallsih HD that is slow.

      The driver can also 'buffer' some of the data out. So if you have a bad app that is writting one byte at a time it may not 'sync' the whole cluster until it has not been touched for a period of time.

      For a windows enviroment it will be more for portable huge amounts of data. There will be some interesting devices coming!!!!

    2. Re:32GB and shrinking... by Overzeetop · · Score: 1

      Yeah, there may be writes, but how much actual data are we talking about? 100k cycles on a 32GB drive is, um, 3.2petabytes of written data. I don't know, but I think even Vista can run in that footprint for quite a while.

      --
      Is it just my observation, or are there way too many stupid people in the world?
    3. Re:32GB and shrinking... by ePharaoh · · Score: 1

      Well, don't know about Windows and it's registry. But it sure would be fun to use on my Linux box, as my root partition.

      Ofcourse, I would still use a normal HDD for /tmp, /home, /var where I "dirty the carpet" often. But my root partition rarely gets modified. Moreover, this is where most installed programs reside. A flash drive would fit-in perfectly.

  51. Environmentally friendly too by failedlogic · · Score: 2, Funny

    I would think the one advantage that Flash drives have over HDds is they're more environmentally friendly (if you don't count the huge packaging they're packed in at retail).

    They are small and lighters and take less space (doesn't use as much fuel to ship), don't produce much heat, use less electricity, and I think there's probably less wasteful throwing out a little stick when its bad than an HDD.

    1. Re:Environmentally friendly too by Anonymous Coward · · Score: 1, Funny

      Unfortunatly, flash drives are made out of the baby seal bits that can't be processed into Prius upholstry.

    2. Re:Environmentally friendly too by failedlogic · · Score: 2, Informative

      Damn, that answer really made me laugh. It was the last answer I ever expect. Is it really true. ;)

      Thanks for the laugh.

  52. flash wear-out by soundofthemoon · · Score: 4, Insightful

    Flash memory cells will indeed wear out after some number of writes. This number is typically pretty high, on the order of a million writes. For most file operations that will probably be a higher MTBF than a magnetic disk with moving parts. Any significant problem would be with hot spots, like VM backing store and file system tables. However you can level wear by using cells in a something like a round-robin fashion. Remember that contiguity isn't an issue with flash because there is no seek time waiting for the head to move. There will probably be some challenges in balancing wear leveling against optimizing file system and VM performance, but in the long run flash drives will likely be much faster and more reliable than magnetic disks.

    1. Re:flash wear-out by networkBoy · · Score: 1

      Wrong about the cycle count, and the wear leveling.
      Cycle count is 100-500K for nand and nor flash.
      wear leveling is already built in to all Flash.
      the leveling is accomplished on a block level. If you want to kill a flash drive compute the block size and write enough data to fill n-1 blocks then continuously update a small file with new data.
      you'll burn away that last block in a few weeks to a month.
      -nB

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    2. Re:flash wear-out by petermgreen · · Score: 1

      i thought they used extra non user accessible space for the wear leveling. If they were really building in understanding of filesystems like you suggest it would be obvious to anyone using an alternate filesystem.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    3. Re:flash wear-out by TheRaven64 · · Score: 4, Informative
      If you want to kill a flash drive compute the block size and write enough data to fill n-1 blocks then continuously update a small file with new data. you'll burn away that last block in a few weeks to a month

      Actually, this won't work. The wear levelling doesn't know if a block is 'full' or not, so it will just switch the contents of a pair of blocks. Your frequently-written file will move all over the flash chip(s), and so will your static files.

      --
      I am TheRaven on Soylent News
    4. Re:flash wear-out by networkBoy · · Score: 1

      Some nor devices have redundant blocks, but in my experience nand devices do not. Also note that the principal still applies if there is a redundant block, you will be swapping data between two blocks and not the rest of the array.
      Usage models indicate that moving the array arround to level wear will cause more overall wear than leaving programmed data alone and simply wear level across unused blocks.
      -nB

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    5. Re:flash wear-out by networkBoy · · Score: 1

      then you and I work on quite different technologies. The wear leveling _should_ know the block's status (program level) as it would need to replicate that data elsewhere.
      -nB

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    6. Re:flash wear-out by petermgreen · · Score: 1

      but when your working below the filesystem level (and as i said if they were building in understanding of the filesystem it would almost certainly be obvious to anyone who formatted one in anything other than its default configuration) how do you know what blocks are unused without having extra blocks above those you make visible to the host?

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    7. Re:flash wear-out by networkBoy · · Score: 2, Informative

      We are talking about different layers, now I understand what you're saying.
      The flash device has a control processor on board that manages the charge pump operation (for writing and erasing) and keeps the status of the array (bad blocks). When the application layer asks to store data the flash control processor goes and stores the data on the next erased block with the least number of erase cycles against it (erase is what causes damage to the flash cell). The application layer has no idea where in the array the data is physically located, nor does it care. The FCP tracks all of this and it was this layer of the device I was refering to. There is not any reason to move data from one block to another, just to free up a block with less erase cycles, as you have no idea whether that data is persistant or not.

      Most flash devices have three arrays: Main array space, microcode array space, and processor(configuration) array space. There may be redundant blocks available, but of these arrays only the main array is useable for storing data.

      YMMV as I work on NOR devices and as such are more closly alligned with memory devices rather than block devices.
      -nB

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    8. Re:flash wear-out by justthinkit · · Score: 1

      It sounds to me like there is a fundamental conflict between how a hard disk is used and how to preserve a flash device -- one is best left empty and the other tends to always be full. It seems absolutely inevitable that no matter the size of the flash drive, it will be mostly or completely full and hasten its own demise in the worst way.

      --
      I come here for the love
    9. Re:flash wear-out by Anonymous Coward · · Score: 1, Interesting
      I work for a company that places core OS components on a compact flash disk plugged into an IDE adaptor. We've mounted it read-write (*nix OS) and have performed the FS optimizations like no-atime.

      They still die with distressing regularity!!!



      Noone in our engineering department has ever given a good answer as to why - my guess is the parent poster is correct - VM and expecially FS tables - inodes can get hammered. At least its the best explanation I've come up with. If anyone can REALLY explain how wear leveling interacts w/ an OS and inode allocation, I'd love to hear it. What happens when an inode is written and the wear leveling algorithm "moves" the inode?

    10. Re:flash wear-out by thrillseeker · · Score: 1
      It seems absolutely inevitable that no matter the size of the flash drive, it will be mostly or completely full and hasten its own demise in the worst way.

      It would make excellent storage for mostly static files such as the operating system and programs - i.e. non-user data. I'm wondering if I could slide it into the floppy slot of my large screen Thinkpad. T'would be a nice upgrade to a system I'm otherwise happy with.

    11. Re:flash wear-out by John+Whitley · · Score: 1

      Flash memory cells will indeed wear out after some number of writes.

      Also note that modern hard drives have various layers of reliability engineered into their design. There's all sorts of bad data at the level of the signal hitting the drive head. That could be spot media failure, a transient failure (e.g. signal magnitude wasn't high enough on that pass) and so forth. Error correcting codes(ECC) and bad block detection and marking are examples of techniques used to achieve the desired levels of reliability. Without these approaches, the storage available on modern drives would be radically lower than it is, since we'd have to rely solely on media and mechanism integrity. That is, to get a net reliability of failures to an acceptable level, a battery of approaches are used beyond media integrity.

      The same general concepts can be applied to a flash-based drive to improve reliability under the assumption that media failures will occur, but should be recoverable. In the case of flash, a write distribution mechanism would also be added to the techniques used to improve reliability.

    12. Re:flash wear-out by Cyberax · · Score: 1

      Just keep physical-to-logical sector mapping table in controller.

    13. Re:flash wear-out by Xiaran · · Score: 1

      You are correct. It is very difficult to destory a flash device. I have destory one in pretty much the way that you describe. I used to have a job writing an encryption driver for the PocketPC platform for SD and Flash cards. I used to perform stress, performance and memory leak tests on the devices by basically hammering read/writes of blocks to the devices. It took me about 3 or 4 months of doing this on a test rack before oen of the SD cards failed :) My estimate(guesstimate really) for that device was approx 200K direct block writes. Most of the devices survived such treatment... but I am sure they were gettign close.

    14. Re:flash wear-out by Anonymous Coward · · Score: 0

      You put your VM on a flash drive when you know it has a limited number of write cycles, why?

    15. Re:flash wear-out by m50d · · Score: 1

      Try using UDF - it has a special mode for dealing with media where you shouldn't rewrite the same sector lots of times (it was pretty much designed for CD-RWs, after all)

      --
      I am trolling
    16. Re:flash wear-out by vhogemann · · Score: 1

      There is already one filesystem that try to address this issues, is the JFFS. Its used by the Familiar Linux distribution, aimed at iPaq handhelds.

      --
      ---- You know how some doctors have the Messiah complex - they need to save the world? You've got the "Rubik's" complex
    17. Re:flash wear-out by MyNameIsEarl · · Score: 1

      Enough of this wearing out stuff, where's my new 32 gb Nano?!

  53. Flash memory that works has a much longer MTBF by Expert+Determination · · Score: 1

    I can't figure out what you're claiming. Are you saying that flash has a longer MTBF as long as you don't defrag it. You appear to be saying that - in which case you're suggesting it has a worse MTBF than hard drives.

    --
    "The White House is not an intelligence-gathering agency," -- Scott McClellan, Whitehouse spokesman.
    1. Re: Flash memory that works has a much longer MTBF by Anonymous Coward · · Score: 0

      The MTBF is relative to how many writes are done on a flash drive. Defragging moves data around and writes data to different places (unneccesary on flash drives), so yes it does reduce the MTBF on flash drives and it is ALSO not needed on flash drives.

    2. Re: Flash memory that works has a much longer MTBF by Expert+Determination · · Score: 1

      Let me rephrase this: I've never heard anyone worry about defragging hard drives reducing their lifespan. But lots of people seem to worry about defragging flash drives. It's trivially obvious that defragging any kind of drive reduces its lifespan. What I want is to compare apples to apples. For the same number of writes do flash drives have a worse MTBF than hard drives? Every time I ask this people find ways to avoid answering the question. One reason I ask is that I seem to go through flash drives pretty fast without ever defragging them.

      --
      "The White House is not an intelligence-gathering agency," -- Scott McClellan, Whitehouse spokesman.
    3. Re: Flash memory that works has a much longer MTBF by zippthorne · · Score: 1

      but similarly, you shouldn't ever even need to defrag a flash drive. Defragging is done on hard disks because of the latency involved in moving the drive heads between reads. IC based memory doesn't have this problem because there is no mechanical step involved in reading a different sector. You can read out all the data in any given addressable unit as quickly as any other addressable unit.

      So the total lack of a specific shortcoming of hard disks which requires lots of writes balances to a certain extent the shorter write-life of flash drives.

      --
      Can you be Even More Awesome?!
    4. Re: Flash memory that works has a much longer MTBF by MrLizardo · · Score: 2, Insightful

      The problem is that you want an apples-to-apples comparison of apples and oranges. The primary reason for hard drive failure is failure of the mechanical moving parts. The primary reason for flash drive failure is destroying a cell by writing to it too many times. Also, your statement about it being "trivially obvious that defragging any kind of drive reduces its lifespan" isn't quite as trivially obvious as you think. A hard drive will almost certainly suffer a mechanical failure long before it's gone through its allotment of spare blocks. On a flash drive that's written to a lot, bad blocks cropping up will probably be the first thing to go wrong.

      If you buy a flash drive, fill it with data, and then never write to it again, you can read all you want and it's minimum MTBF will be ~10 years (AFAIK, there's no reason they couldn't last longer, it's just that more testing needs to be done to prove that they will last longer).

      Another problem in comparing hard drives and flash drives is based on what kind of environment they're subjected to. Flash drives are usually portable devices that live in pockets, and are subject to static shocks and being plugged/unplugged on a regular basis. hard drives for the most part live in computers where they're protected from the elements and aren't often disconnected, especially not with the power on. In your case, I'd be willing to bet that your flash drives are dying from a failure in the onboard controller (rather than individual cells dying). It might be interesting to purchase a small USB hard drive and compare how long it lasts when subjected to the same environment as your flash drives.

      -Mr. Lizard

      --
      ^I'm with stupid.^
  54. Forget using this for a Microsoft Windows OS by DigiShaman · · Score: 0

    The registry hive is ALWAYS writing the drive. In fact, it's a streaming write of data. Kiss the MTBF of these flash drives goodbye!

    --
    Life is not for the lazy.
    1. Re:Forget using this for a Microsoft Windows OS by Dibblah · · Score: 1

      I'm afraid I have to call FUD.

        Directory of C:\WINDOWS\system32\config ...
      09/03/2006 14:38 36,700,160 software
      09/03/2006 14:38 262,144 SAM
      09/03/2006 14:38 524,288 default
      17/03/2006 11:12 27,525,120 system ...

      ie the last write to the system hive was 4 days ago. Not quite within my definition of "streaming". And yes, the write times on these files are valid even though they're held open by Windows.

    2. Re:Forget using this for a Microsoft Windows OS by DigiShaman · · Score: 1

      I call BS. Not saying you're lieing, but rather Windows is giving you false information.

      Try this. Try installing any program that writes to the registry (most if not all windows program does). After you've done this, try resizing windows in various shapes and configurations (this too is stored in registry). Give it about a minute at the most....now yank the power from the PC.

      Turn the PC back on. You might notice the window framing is the same for the folder in which you resized. Also, you will not have to reinstall the application.

      --
      Life is not for the lazy.
    3. Re:Forget using this for a Microsoft Windows OS by kylef · · Score: 1
      Try this. Try installing any program that writes to the registry (most if not all windows program does). After you've done this, try resizing windows in various shapes and configurations (this too is stored in registry). Give it about a minute at the most....now yank the power from the PC.

      Window sizing policy is entirely application-dependent. Windows itself does not record window size or position information at all. Applications are free to persist this information however they see fit. Personally, if I chose to use a registry value to persist this information, I would only update the value at program exit, and not at every window resize/position event. Anything else would be poor application behavior on my part.

      Regardless, this is not a Windows OS problem and could happen on Mac OS X or Linux just the same. Just replace "registry" with "application configuration file" and you have a similar situation. The problem lies with poorly designed window metadata persistence, not OS-level policy.

    4. Re:Forget using this for a Microsoft Windows OS by Shanep · · Score: 1

      I'm afraid I have to call FUD.

      Take a look at SysInternals tools, they have a tool to watch registry activity. Start it up and watch the lines of reads and writes quickly run off the screen. Lots and lots of activity ranging from steady activity to quick bulk bursts. You REALLY need to use the filter when using their registry watcher, because what you are interested in, will be drowned out by the flood of activity. Even when using the filter, depending on the application you are watching, there can be lots of activity.

      I'm thinking that this activity is mostly confined to activity in RAM, with the registry cached and then changes are commited to disk perhaps at regular intervals and when someone logs out or shuts down. If your numbers are correct, then I can only assume that not every write to the registry is going to cause a timestamp update on the registry files themselves. Because the activity, including writes, was VERY high when I looked.

      --
      War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
    5. Re:Forget using this for a Microsoft Windows OS by Dibblah · · Score: 1

      This behavior changed with Windows XP. The registry is now _only_ committed when an application asks Windows to commit, or on shutdown.

  55. price? by spartacus_prime · · Score: 0

    At $6400, who will buy the 32 gig one? While I love the extra space, it seems a little unecessary. You could use an iPod for the same purpose (external hard drive) for a hell of a lot less money.

    --
    If you can read this, it means that I bothered to log in.
  56. They were not USB drives, by Dr.+Evil · · Score: 1

    They were isolinear chips.

  57. Yes please by fletchzip · · Score: 0

    2 x 32GB RAID 0 Laptop, put me down for one.

  58. Old news? by Anonymous Coward · · Score: 0
  59. Thanks for link by Anonymous Coward · · Score: 0

    interesting site.

    Yes, the form factor is different but I find it interesting that these are built to mil spec. Probably means the US military is already deploying these in the field.....and now we can to.

  60. So doing the math... by Maxo-Texas · · Score: 1

    $1000 next year.
    $500 in two years.
    $250 in three years.
    $125 in four years .. getting VERY interesting here...
    $63 in five years.

    How I would design my laptop that used this...

    1) put in multiple standard SD plugs.
    2) have builtin copy function to backup card to another card.

    Easy to upgrade as the ram gets larger and cheaper.

    I thougth SD ram was small- but a guy at work got a memory card for his cell phone (something like 512mb to hold mp3's) that has an SD card ADAPTER that this teeny cell phone mp3 memory slides into. The darn thing is about .25" x .375" x almost 0 thickness. That kinda blew me away.

    --
    She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
    1. Re:So doing the math... by Anonymous Coward · · Score: 0

      Problem is, the price of NAND based flash doesn't always drop that fast, and eventually as you make 'em larger and stuff, you get less on a die, and yields go down too, etc. I'm not sure if you can rely on it going down in price like that.

      And regardless, it's still overpriced even if it does. Because HD prices are falling at least as fast - not in absolute price, but in terms of how much GB/$. We're around 40 cents/GB now (300GB for 120$). In 5 years time, you'll be getting a LOT more for that same price tag.

      And not only it'll still be overpriced, it'll be mostly useless for most people. I just bought four 300GB drives this week. I was buying 40GB drives 5 years ago. So waiting for them 5 years to be affordable is still 10 years behind - more or less eternity in the computer world. In 5 years, we'll be buying drives over 1TB for 200$, so tiny little 32GB'ers...

      We need more and more storage nowadays. People start having music/movie/ebook/photo/pr0n/whatever collections. I've skipped the terabyte a few years ago already. This thing is already too small for my mp3s, or my photos, or anything really (heck, the kids' movie collection alone is over 1TB, and that's in so-so mpeg4 asp).

      I would be real nice to have super fast HDs for cheap, but if they're that tiny then I got no use for it...

    2. Re:So doing the math... by matt21811 · · Score: 1

      " $1000 next year.
      $500 in two years.
      $250 in three years.
      $125 in four years .. getting VERY interesting here...
      $63 in five years."

      This assumes that 32 Gig wont be considered some trivial amount of storage in 5 years from now.
      It assumes also that flash will continue to get cheaper at the rate of 100% per year. Which it might.

    3. Re:So doing the math... by the+grace+of+R'hllor · · Score: 1

      That would be mini-SD. I currently have a card like that which holds 2GB. Bought it on eBay.

      There is also microSD, or Transflash, which is smaller yet again (about 50-75%?) and weighs in at a maximum of 1GB at the moment.

    4. Re:So doing the math... by Zaatxe · · Score: 1

      This math seldom works for electronics and computers. I remember when I were at college (about 1995) I used to read that computers should reach 300 MHz by the year 2000. If my memory doesn't fail (well, I'm getting old), by the year 2000 the GHz barrier was broken. 6 years later, my 1-year-old notebook runs at 2GHz and my girlfriend's 6-month-old notebook at 1.5 GHz. And I don't see many desktop for announced that goes much further from 3 GHz.

      MORAL: forget first degree equations when you are talking about computers.

      --
      So say we all
    5. Re:So doing the math... by Maxo-Texas · · Score: 1

      That was it-- it was a transflash. It's too small in my opinion! Almost need a tool of some kind to hold it to put in the slot.

      --
      She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
    6. Re:So doing the math... by Maxo-Texas · · Score: 1

      Yea- I noticed that on processor speed- we should have been at about 10ghz now and instead we seem stalled at 3ghz with some 4ghz/dual processors.

      I remember in the 90's you could never get in at the right time, because almost as soon as you bought it, a clearly faster version would come out.

      --
      She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
  61. Yes but... by Anonymous Coward · · Score: 0

    The isolinear chips were connected using the USB 45.1 specification - light years ahead of the antiquated USB 2.0 specification. The data is actually sent at WARP speeds and gets to its destination before you actually send the data.

  62. Prices could be lower by babakm · · Score: 1

    If I'm reading the lastest NAND spot prices correctly, it costs ~$20/gigabyte after the recent drop. That would put a 32gb drive at $640+cost of other components and assembly. I'd safely guess that we're actually talking about less than $500 for the drive when it hits the market.

  63. Flash Drives vs. HD by punkguitarist · · Score: 1

    I read an article in The Economist about Flash Drives vs. Hard Drives. The rate of expansion of them isn't closing much, if at all (meaning everytime Flash gets bigger, so does the HD), so it is hard to see it ever dominating. Although there are many other more current technologys (very much like how NAND is superior to NOR flash memory) which can hold much more information, which will likely become more mainstream in the near future.

    1. Re:Flash Drives vs. HD by matt21811 · · Score: 2, Interesting

      Can you provide the link for the Economist article. This is an area of interest for me.
      My own research shows the opposite is happening. Flash is charging hard after disk and the rate it is catching up is accelerating.
      http://www.mattscomputertrends.com/flashvsharddisk .html
      I am due to update this years figures but a quick analysis shows the trend is continuing.

    2. Re:Flash Drives vs. HD by punkguitarist · · Score: 1

      Yeah sure, It is on the economists website: http://www.economist.com/displaystory.cfm?story_id =E1_VVSTVQQ - you need a subscription to view the article, but you can get a day pass by watching an advertisment on their homepage [http://www.economist.com/%5D. The article is titled "Not just a flash in the pan" - it appeared in their technology quarterly a couple of weeks ago. Good read with many pictures and graphs to illistrate the trends. "And as the price per gigabyte of both technologies falls, the fight shifts to ever higher capacities"

  64. Why have a hard drive at all? by aphaenogaster · · Score: 1

    The real future of mobile computing without worry for battery power is the laptop equivalent of a sunray. All the damn thing should have is a wireless card a 100mhz processor, a kick butt video card, and everything else should be on a big ass server in some airconditioned room somewhere. I imagine an apple newton looking thing.... or a compaq m300 without half the stuffing. Hauling a harddrive around is just asking for data loss and suffering. Particularly if the harddrive cost you 1000 dollars. When was the last time you brought your laptop somewhere that did not have wireless/wire? Excluding games and coach airplane rides.

  65. Laptop Storage? by Brainfuck+R00lz · · Score: 0

    Am I alone in thinking that 32gb is still a bit too small to be a Hard drive on a Laptop. Windows now takes up nearly 4gb of space, leaving 28gb for programs and data, most people use laptops like they use desktops and never get rid of data, that isn't a whole lot of room to play when programs now get over a gig. Now just imagine in a few years when vista is out how well is 32gb going to hold up? It's going to take atleast twice that to get me to buy a computer with a flash drive, but on the up side it helps out alot for POS and other applications where energy consumption is an issue.

    1. Re:Laptop Storage? by Anonymous Coward · · Score: 0

      Windows is made a lot smaller when you 1) Dont copy the CD to the hard drive (though the install does this for you, you can remove it) and 2) Dont set up the swap file to be 1.5x your RAM (aka install enough RAM to replace your swap file).

      I've gotten XP to run in smaller spaces than 2 GB, though no where near as tiny as Linux ;) Usually, it seems a default install of newer Windows is 600 mb (oddly, the size of the CD it came on, wow) + 600 mb for the copy of itself it puts on the hd + 1.5x RAM for swapfile. And try not to install EVERY trinket that comes with the OS...

    2. Re:Laptop Storage? by geekoid · · Score: 2, Interesting

      IT would be large enough for business.
      Is windows is large and so is office, but thats 10G. The remainder is emails and docs, which don't take a lot of space.

      Now, you add movies, mp3, games, etc . . . it won't be big enough.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    3. Re:Laptop Storage? by brown-eyed+slug · · Score: 1

      It's not beyond the manufacturers to include multiple drives. Only a question of cost as I see it.

    4. Re:Laptop Storage? by Brainfuck+R00lz · · Score: 1

      have you ever checked the average person's work computer, most people never delete e-mails, docs, have pictures and music on there too, 32gb really isn't alot of room for most people, they've become accustom to having 160gb of space, they could care less that thier laptop has a 7 hour battery life and weighs only a pound and a half. It works in theory, but in practice it wouldn't hold up. now, as a pocket drive it would be great, i would love to have something that small with that much room, it could also be adopted to MP3 Players too, I jsut can't see it as a feasable replacement for a hard drive just yet, even in buisness markets, even budget laptops now have 60gb, I think thats really the magic number when it comes to bringing in a new tech for main storage (until programs get bigger and the pace of change increases yada yada etc.)

  66. Data gets laid in that episode by Nonillion · · Score: 1

    I have this episode on VHS, this is the one where Data gets to bone that hot security officer.

    --
    "I bow to no man" - Riddick
    1. Re:Data gets laid in that episode by Anonymous Coward · · Score: 0

      Y'ar, she be a tasty one, methinks!

    2. Re:Data gets laid in that episode by Anonymous Coward · · Score: 0

      Hey, that hot security officer was somebody's daughter!

  67. Another level of cache by prozac79 · · Score: 2, Insightful

    So, I see a lot of "But my hard drive stores 500 GB at a fraction of the price" comments. However, a flash drive can be yet another level of caching that sits between memory and the hard drive. The order of data access would then become L* cache, RAM, flash drive, hard drive. 32 GB is plenty of space to load the OS and run normal apps like a web browser, email client, etc. So, instead of writing a page/swap file out to the hard drive, one would be able to write it out to the flash drive instead. This would result in faster reads and not consume as much power (think laptops). Also, since it's persistent (unlike RAM) then you could have better computer boot times. Basically the mechanical hard drive becomes a type of nearline storage device that gets accessed later (and less often) in the pipeline. Does that make any sense? I often fell asleep in my OS class in college.

    --
    "Oh dear, she's stuck in an infinite loop and he's an idiot" -Prof. Farnsworth (Futurama)
    1. Re:Another level of cache by Anonymous Coward · · Score: 0

      I remember reading some article on /. - was something about putting few hundred mb of NAND flash (or whatever) as a cache inside the hard disk. So writing to disk would be more efficient. I spose they had some way to speed up loading too. Was supposed to save battery life too.

    2. Re:Another level of cache by maxume · · Score: 1

      Swap? Currently, at Newegg, flash is about 3x the price of (s)dram. At that rate, you better really, really, need that swap space more than you need ram.

      --
      Nerd rage is the funniest rage.
    3. Re:Another level of cache by Avast+Yee · · Score: 2, Interesting
      Also, since it's persistent (unlike RAM) then you could have better computer boot times. Basically the mechanical hard drive becomes a type of nearline storage device that gets accessed later (and less often) in the pipeline. Does that make any sense? I often fell asleep in my OS class in college.


      Actually, I remember reading some article years back about the "future of OS storage" or some such thing. I've often thought about how it would be much nicer to have your OS stored on an intermediate storage device between your primary memory and your hard drive, at least for people who don't update their OS very often. I don't know how plausible a system like this would be, but I think there would be a lot of advantages. If you could have fast SSD memory that you could put your OS on and use your hard drive for temporary files, multimedia, and whatever else, your boot time would be considerably less. Further, a physical switch of some sort that you flip on the front of your computer once you have the OS installed and configured could make the OS storage read only, improving its security from viruses and things, couldn't you? Most of the OS is not changed very often, so it seems ideal to me. You wouldn't even have to have a very large device; five or ten gigabytes would suffice. Are there systems like this? If not, why?
  68. FlashBelt - not a bad idea by drewzhrodague · · Score: 1

    It actually makes sense to have a wearable storage facility. We do have plenty of gizmos in our pockets, our bags, and some of us have them on our heads. Why not use your existing headset to play MP3s from your wearable storage? Dump a recording of that phonecall to disk. Record video, log biometric data, log your GPS data, Wi-Fi signals, timestamps, security tokens -- all of that. Why does each device need it's own flash? You could netboot them all.

    While you're at it, I think that any device signifigantly large enough to handle sharing any kind of volume over Bluetooth or Wi-Fi, you might as well include USB-Master support, a couple of i2c interfaces for the thermistors and accelerometers, and probably support for extra batteries, and optionally a simple display format -- NTSC or simpler. I don't think it would signifigantly increase the package size to add those things, except for the connectors themselves. I'm assuming wearable as in wearable under-the-clothes and discrete wearable computing.

    --
    Zhrodague.net - I do projects and stuff too.
    1. Re:FlashBelt - not a bad idea by Doc+Ruby · · Score: 1

      I like a belt because they're already worn by most adults anyway, they're the optimal weight distribution for carrying, don't need washing, and Batman already proved their utility.

      I like Bluetooth because it lets the belt contain bulk without a UI. Displays and other multimedia UI components are best made separate components that clip to the belt for recharging, with Bluetooth offering the same integration benefits as a single device, but with more focused operation for different uses. Once Toshiba et al produce cheap fuelcells the size/weight of a belt buckle, liquid fuel will offer longer sessions through better energy:volume than batteries, and perhaps even liquid transfer for immediate "recharging".

      --

      --
      make install -not war

    2. Re:FlashBelt - not a bad idea by fatcatman · · Score: 1

      It actually makes sense to have a wearable storage facility.

      Uhm... Cellular phone... memory card... bluetooth built in. It's all here. All you're missing is the interoperability issue. IE, why can't my MP3 player pull data off my phone's CF card via bluetooth? There's no technical reason it can't. Just some software, and bluetooth in the MP3 player...

    3. Re:FlashBelt - not a bad idea by EZLeeAmused · · Score: 1
      I don't think it would signifigantly increase the package size

      What do you mean? This might be the first thing you can get on a computer that really does increase package size.


      Sorry, I'm usually better at resisting base humor.

      --
      Some see the vessel as half full; others see it as half-empty; We pour it out on the floor and laugh
    4. Re:FlashBelt - not a bad idea by Anonymous Coward · · Score: 0

      .... and why can't i play mp3s directly on my phone, and then automagically put the music on pause when i make or recieve a call? oh wait....

  69. I bet Microsoft is giddy over this one. by Rifter13 · · Score: 1

    They no longer need to sell software. They sell a little hard drive with Windows (SU) installed. It plugs into the computer, it goes out, and configures itself to the system. So, you have a computer that boots faster. This would be your "boot device". Kind of cool, but I could see problems. Only MS could get to the drive, so other OSes have a harder-time in a multi-boot scenerio. Since your OS and your drive are one, you want to upgrade your computer? Nope! You need a new OS disk to put into a new computer. Sorry. Some cool stuff could happen from this development... but some very annoying things could crop up as well.

    1. Re:I bet Microsoft is giddy over this one. by Vo0k · · Score: 1

      This got me thinking.
      Wow.
      3 seconds from PowerUp to BSOD.
      Just wow.

      --
      Anagram("United States of America") == "Dine out, taste a Mac, fries"
  70. Shock and vibration by Intron · · Score: 2, Interesting

    The article didn't mention shock and vibration resistance, but the flash is likely to be far more rugged than a rotating drive. Might have better temperature specs, too. Once we get flexible flat screen displays, I'll be able to drop my laptop without having a heart attack.

    --
    Intron: the portion of DNA which expresses nothing useful.
    1. Re:Shock and vibration by DeathElk · · Score: 1

      CRASH!! Oops! Doopdi-doopdi doo, doopdi-doopdi doo.. CRASH!! Oops! Doopdi-doopdi doo, doopdi-doopdi doo.. CRASH!! Oops! Sorry...

  71. Raptors have their place. by Anonymous Coward · · Score: 0

    anyone who needs scsi like performance and relilabity in their system.

    That combined with the fact that the raptor is the ONLY 10k sata drive (due to the fact that WD dosn't have a scsi division, therefore they don't have to worry about losing profit from an existing division) in existance.

  72. NAND Flash HDD by Anonymous Coward · · Score: 1, Informative

    Samsung announced these drives last year and said that they should be available by 8-05. I have been looking regularly to see when they are going to hit the market for close to a year now. I wish they would go ahead and release them or plan announcements a little better.
    If they could get the price down to near $10/GB I would buy one. I don't store mass amounts of audio or video on my laptop so a 16 or 32 GB would be fine for me. The decrease in power consumption and the increase in battery life could make the investment well worth the cost. I can currently get about 5.5 hours of battery life out of my laptop, if this could be increased to 7 or more from the reduced power consumption I would be all for it.

    http://www.samsung.com/PressCenter/PressRelease/Pr essRelease.asp?seq=20050523_0000123980

  73. Proven Reliable. by Yiliar · · Score: 1
    I have been testing some 2GB Compact Flash modules for use in embedded Linux systems.

    Using Mysql, 25,000,000 cache flushes (yes, 25 MILLION) were forced to disk with not one error! It took a few weeks of pounding the database with 1000 inserts and 1000 deletes at a time. By keeping the number of 'extents' constant, we observed the same inode referenced each write. The count of transactions was in the Billions, but we were only interested in the number of writes, which we watched using tools like vmstat. We averaged about 11 writes per second.

    Each generation of FLASH memory seems to get better, and my personal confidence in FLASH for primary OS platforms is rock solid.

    That test system has been on and off for several months now with the same FLASH, without changing or updating the OS, and still not a single read or write error reported by the system.

    1. Re:Proven Reliable. by l33tmike · · Score: 1

      But what is the reported size of said '2GB' compactflash drive now?

      Reading above comments suggest that as memory locations use up their write limitations, they get flagged as dead and a different location is used instead.
      This is all without the user knowing about it... possibly even without the OS knowing - so you may not have got any errors logged.

      Please prove me wrong with this theory and say that the compactflash drive is still at full capacity - this new, larger solid state technology is looking very promising!

      Mike

  74. Re:95%? Where? by Anonymous Coward · · Score: 0

    GP means the harddrive only accounts for 15% of the total system's consumption, the rest going to the backlight, cpu, etc.

    Basically he's saying the harddrive doesn't consume that much, and dimming the backlight would do more to improve your battery life.

  75. Magnetic drives are superior! by Drunkulus · · Score: 0

    And records sound better than CDs
    Tube amps rule
    Carburetors, forget chipping
    and the New Testament is for sissies.

    1. Re:Magnetic drives are superior! by Anonymous Coward · · Score: 0

      ... And 35mm print film has a better exporsure latitude than a digital sensor.

      oh, hang on, that one is non-negotiable.

    2. Re:Magnetic drives are superior! by Anonymous Coward · · Score: 0

      And tube amps are better than solid state amps!

  76. Why a laptop? by Geekbot · · Score: 1

    If I could have a 1.8 32GB flash drive I wouldn't need a laptop. I could just use handheld for anything that I would use a laptop for.

  77. $700 is actually cheap by sk8fool · · Score: 1

    Try looking for a high speed high memory pen drive now and you'll spend close to $800 for an 8gb drive

    1. Re:$700 is actually cheap by Anonymous Coward · · Score: 0

      no. newegg has them for $320.00

  78. so, solid state technology by yagu · · Score: 1

    So, solid state technology can now provide us with a storage device large enough to hold Office.

  79. Re:I bet Microsoft is giddy over this one. OS? by davidsyes · · Score: 1

    Open Source, too? Imagine the laptop manufacturers/assemblers, too. They could save on the need to install the obligatory hard drive.

    Now, if they could get virtual OS testing done on their hardware, they could reduce the size of or eliminate fans.

    Then, if the LT mfrs don't need to ship a hard drive, the weight goes down, too. Then, the form factor can change a bit. That space can be filled with more peripherals, maybe a peripherals dock or insert, where *nuxs can be all-system on the 32 GB disk/stick and data on another removable block; *doze an be drive C:\ on the 32 GB disk/stick, and if there are complaints about vista gobbling up disk space, then that can be a punishment for ms and their bs disk-gobbling campaign. Drive D:\ can be, as in the case of the *nuxs, the data space.

    Having the OS and the data forced to be separate due to physical size/memory/storage constraints of the media could be a GOOD thing. Devices can be made smaller, lighter, and less power-hungry. Even if the LCD IS taking some 25-35% of the battery power, having an OS with few writes to its own area, and having users become used to occasionally copying their data to offline storage might be an interesting change in the industry. Now, if the DVD spinning could be dispensed with by having an nutating or wide-scan optical reader... (burning could still need a motor, I suppose...)

    Now, to keep ms from creating another anti-trust issue, this time over the access to and pricing of the blank disks... I could just see it now... (sycophantic, shrill ms lawyers and marketing) "We must deter piracy and illegal license us of disks product activation codes. ..." ala the blank CD/DVD pricing/padding to combat music & movie "piracy"...

    --
    Previously: "Linux... Toward the Sunrise..." Now: "Linux... Toward the-- No, now, part of Every Sunrise"
  80. The solution is coming... by Anonymous Coward · · Score: 0

    The Domodrive is coming soon. www.movai.com

  81. New File System Optimized for Flash? by Anonymous Coward · · Score: 0

    I dont know enough about how filesystems work, but are current filesystems (be it NTFS or ext3) particular to using discs over flash? Of course they work - I use FAT on all my little flash cards - but would there be enough a reason to redesign a file system from the ground up to work for flash? Or are filesystems high level enough to where this isn't a worry, and the real change should be on the integrated electronics for the SDD.

    1. Re:New File System Optimized for Flash? by Vo0k · · Score: 2, Interesting

      Most likely it will (and usually is) made on the low level of the drive electronics - sectors as in commands sent over the tape don't map to specific bits in specific chips but are dynamically assigned and rotated, so that FAT while still appearing to be in the same place as always for the OS and disk controller (on motherboard) in fact migrates thorough the physical drive memory being dynamically relocated by the drive logic to new areas, so that no single chip gets unfairly high number of writes leading to busting the memory. This is completely transparent to all the hardware and software outside the drive, except maybe for undelete utilities.

      --
      Anagram("United States of America") == "Dine out, taste a Mac, fries"
  82. Speed ~ 1/Time by douglips · · Score: 1
    Read speed: twice as much is better.
    Read time: Twice as much is worse.

    From TFA:
    According to Samsung, the SSD will read and write data at 57 MB/s and 32 MB/s, respectively. We will have to benchmark such a drive in our test lab to verify this claim but if correct, the Flash disk would be about twice as fast as the latest 1.8" hard drive generation, which was measured at a read speed of 24 MB/s by the engineers of Tom's Hardware. The acceleration is most likely not enough to enable instant-on computers, but we would expect Windows computers to cut the system boot time at least in half.

  83. To increase the speed... by spoop · · Score: 1

    I assume this drive is not one chip, but many? So samsung could break it into 4 separate parts and have a 4 drive raid 0 array in a single drive to increase read/write speeds.

    --
    I blame geof's speakers.
  84. No more altitude limitation on laptop by hdd · · Score: 1

    as oppose to the -200 to 10,000 feet limit on regular hdds

    --
    This Sig is removed due to factual inaccuracy
    1. Re:No more altitude limitation on laptop by Anonymous Coward · · Score: 0

      It has nothing to do with altitude and everything to do with pressure. I doubt people bring their notebook mountain climbing and in pressurized cabins it doesn't make a bit of difference if your 32,000 miles above the ground because the atmosphere within is ~ 1 atm which is what the disk drive is designed to operate at. Totally worthless improvement for 99.999999999999% of lusers.

    2. Re:No more altitude limitation on laptop by miro+f · · Score: 1

      Excellent, now I can brin my laptop with me while scaling Everest and deep sea diving!

      --
      being vague is almost as cool as doing that other thing...
  85. Seek time by tuxlove · · Score: 2, Interesting

    The article says nothing about seek time... Obviously, there is no seek time with a flash drive. Accessing memory is the same cost, regardless of the address being accessed. This presents a potentially massive performance improvement over traditional drives, transfer rate notwithstanding. To me, this is the big win.

    If flash drives were more commonplace, it would revolutionize filesystem and database development. No longer would you have to care about sequential access, keeping blocks contiguous, etc. This would change everything. I'm amazed that you don't hear more about this.

  86. 32 MB powerbook drive replacement by sxtxixtxcxh · · Score: 1

    some guy hardware hacked a powerbook, replacing the internal hard drive with a 32MB compact flash card. :)

    now to find someone who's done that with titanium powerbooks and larger flash cards :D

    --
    for a minute there, i lost myself...
  87. 32 GB Flash Storage Drive Announced by infiniphonic · · Score: 1

    Remember that technologies improve.The flash memory we have today that can take 100,000 writes might not be the flash memory of tomorrow.My first (real) hard drive was a 10gig maxtor that cost around 70.00.Think about the technological improvements and price per gig that we have seen in just the past five years.In another five years we might see (hopefully) just that kind of improvement in flash technology.

    --
    Crisis is the rule, not the exception.
  88. Old stuff by threedognit3 · · Score: 1

    Much of the stuff hitting the market today is old hat. Why... It was commissioned by the DoD many years back. What you're seeting today is the result of DoD (and others) requirements for a network centric environment. Many of the innovations you're seeing today is the R&D of the DoD years back. It's not new but it's been proven. Many companies who do DoD work ask that once the technology is well past innovative and no longer useful, they are able to market it to the public. What is available now is the remains of things no longer useable to the DoD. We, the public, are well behind the technology curve. Determining which company came up with the technology and will market it...is the choice...er...chance.

  89. Will there be a heat issue? by Tsaot · · Score: 1

    My main concern is whether there will be a heat issue or not. I regularly run Slax off of my thumb-drive and the heat the thing gives off after about 10 min is unbearable (but it makes a good handwarmer in the winter). If this is what I get from 1 gig, what can I expect from 32?

    1. Re:Will there be a heat issue? by Vo0k · · Score: 1

      your thumb-drive is broken.

      --
      Anagram("United States of America") == "Dine out, taste a Mac, fries"
  90. Music recording by buswolley · · Score: 1

    recording music would benefit from a faster write and read times of a flash drive.

    --

    A Good Troll is better than a Bad Human.

    1. Re:Music recording by nmg196 · · Score: 1

      Why? Music is very low data rate. It's also a constant stream meaning access times are totally irrelevant. I can't see why faster read and write times would help at all. High definition video maybe, but certainly not music.

    2. Re:Music recording by buswolley · · Score: 1

      Even recording 16-24 tracks simultaneously at 24 bit /96?

      --

      A Good Troll is better than a Bad Human.

  91. Probably by Anonymous Coward · · Score: 0

    Well, it matters for malloc() implementations, and my RAM doesn't have moving heads, either.

  92. Caching the write once/ read many things... by thedletterman · · Score: 1
    A few years ago, when flash memory hit the 512MB capacity, and started being viewed as a replacement to cd-rw, I made the observation that if a Mobo manufacturer ever put a 8gb or greater chip on the motherboard to cache the operating system, and commonly used installed programs, they could do great things in terms of reduced energy, heat, and improved latency.

    Give me a bootable NAND C drive built into my mobo damnit! I've been waiting for YEARS! 32 GB is great and all, but i don't need anything over 10GB.. anything over that is rewritable crap that I'd rather store magnetic.

    --
    Any fool can criticise, condemn, and complain, and most fools do. - Benjamin Franklin
  93. bitmicro doing for years by seb42 · · Score: 1

    http://www.bitmicro.com/ have 2.5-inch with sizes of 512 MB to 73.7 GB , but I read that samsung will be the first for the mass market. These solid state storage devices, with there faster access times,can be great for improving mail and database performace without out having to do any upgrades.

  94. This is one looks more interesting to me... by Anonymous Coward · · Score: 0

    1TB Non-Volatile Quantum STORAGE ATA IDE

    http://atomchip.com/_wsn/page3.html (scroll to the bottom of the page)

    Spec:

    DC Input Voltage: 5.0V ±10%
    Standby: 0.5microA
    Read: 1.5 microA
    Write: 2.5 microA
    Reading time:60 ns
    Update time:120 ns
    Temperature Operating: -50+125C
    Humidity Operating: 5 - 95%
    Shock Operating [max]:2,000G

    OK the price is maybe too high for average user but... who knows in the near future when the
    techonology will "get old" enough... is should drop for sure

  95. Speed is disappointing by hcdejong · · Score: 1

    read speeds more than twice that of a normal hard drive

    and FTA:
    According to Samsung, the SSD will read and write data at 57 MB/s and 32 MB/s, respectively...
    the Flash disk would be about twice as fast as the latest 1.8" hard drive generation, which was measured at a read speed of 24 MB/s by the engineers of Tom's Hardware.


    57 MB/s is in the ballpark for the read speed of a 3,5" 7200 rpm drive like the Seagate Barracuda 7200.8.

    Being used to RAM access times being mesaured in ns, this is rather underwhelming.

  96. Really huge win is 0 seek and 0 rotational delay by bdolan · · Score: 1

    This drive should be an absolute success. Who cares about 2 x transfer, by itself ... the seek time is almost always more than time to transfer a large block of date. Plus, the time alone to rotate to the block's location is a huge amount also. At 5400 rpm, 1/2 of a rotation is 11ms, more than the time to transfer almost any block of data. The transfer speed is only good on hard drives when the head is positioned and the disk rotates into position--a 20-50% duty cycle. The flash drive can start transferring immediately and keep transferring--a huge win. Why can't Samsung tell us right now that this runs wickedly fast, or is their transfer rate really some effective speed accounting for the lack of seeks, otherwise we should be seeing 5x plus speeds. This is a trivial benchmark, but if they can make the specs without some odd adjustment this we should have a incredibly productive laptop in a lot of heavy disk based applications--it even would be able to go to sleep way faster. A large premium would be justified easier without any more hesitation for cost.

  97. Gates to mock Samsung? by dtsazza · · Score: 1

    I find this especially interesting after Bill Gates mocked the $100 laptop for, among other things, "hav[ing] it be something without a disk...". Reading the specs of the laptop, it comes with 500Mb non-volatile RAM; so it does have persistant storage, it's just that storage comes in a different form to a hard drive.

    Now Samsung are suggesting that other (all?) laptops "be something without a disk" - is Gates going to mock them too?

    --
    My, that was a yummy potato!
  98. JFFS by mu22le · · Score: 1

    There are filesystems specifically developed for solid state storage, have a look at:

    JFFS2 and YAFFS

    They are being abandoned anyway since most flash drives have built in levelling system (this menans that almost any filesystem is ok).

    OPIE, a linux distro for handhelds device, use JFFS2 for SecureDigital cards .

  99. Sucks by mattcasters · · Score: 1


    Like the conversation between the hair-dryer and the vacuum-cleaner:

    Haidryer : You suck!
    Vacuum-cleaner: You blow!

    --
    News about the Kettle Open Source project: on my blog
  100. This will be really fun... by EmagGeek · · Score: 1

    especially when you start reaching the write-cycle life of the FLASH. Hopefully it at least reallocates data and dynamically shrinks the size of the disk in a way that Operating Systems can understand. Most O/Ses expect the partition table to be constant, but if you start running out of cells and the disk starts shrinking, the partition will also have to shrink with it and the OS will need to be aware of that fact.

  101. Raid and os! by Anonymous Coward · · Score: 0

    i need five 5gig ssd's please. i wanna raid5 them
    use it to only install the os. for the rest i'll
    just go with a 350 gig maxtor thanx u!

    k okay i'm not swimming in money, so maybe
    two in raid1 (but just for the os!).

  102. Re:Really huge win is 0 seek and 0 rotational dela by Anonymous Coward · · Score: 0

    Check your math. 1/2 rev on a 5400 RPM drive is 5.5 ms, not 11 ms.

  103. yes... but... by 1800maxim · · Score: 1

    the real question is not price, the real question is, Will it run Linux?

  104. It's not really that many. by brunes69 · · Score: 1

    I have files in /var/log that are modified at least 5000 times a day. That's only 200 days wth a million writes.

    Granted, load balencing takes care of some of this, but I don't see how you could have any logging whatsoever on your laptop with a drive like this. And for those of us who use laptops for development, heavy logging is a necessity.

    I'm not going to pay hundreds of dollars for a drive guarenteed to fail in a year.

  105. This is an annuity for Storage Vendors by macz · · Score: 1
    If you take in to account the limited number of writes that Flash memory has typically been encumbered with (1 million writes? Any upper limit is a problem) this is the perfect disposable device for storage vendors to saddle customers with. Software can track the number of writes (or even an LED on the drive itself), and when you get in the red, you will have to purchase a whole new unit or lose the ability to reliably migrate off of the old one.


    I don't see the incentive for them to fix this either. Once you get on a fast, low power drive, you won't want to go back to spinning disks... even if they can run for years reading and writing without problems.

    --
    ...But I digress. TREMBLE PUNY HUMANS!ONE DAY MY SPECIES WILL DESTROY YOU ALL!