Slashdot Mirror


Intel Confirms It Will Ship 160GB Flash Drives

Lucas123 writes "Intel has confirmed plans to ship a new line of solid-state drives for laptop and notebook PCs with storage capacities of 80GB to 160GB. While it did not lock in a ship date, Intel told Computerworld that the drives would be available in the second quarter. From the story: 'An aggressive move into the laptop and PC notebook flash disk drive business would catapult Intel into direct competition with hard drive manufacturers such as Toshiba Corp. and Samsung Electronics Co. that are trying to spark demand before their SATA-based offerings are released in the coming months.'"

228 comments

  1. Hmm.. by Anonymous Coward · · Score: 0, Funny

    I call your mother the "160GB flash drive." That can't be just a co-incidence.

  2. Great. I buy a 160GB iPod and now they by crovira · · Score: 1

    tell me I could have got a solid state one.

    Oh well. I'll just have to wait until the moving parts on this one stop moving.

    --
    MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
    1. Re:Great. I buy a 160GB iPod and now they by Anonymous Coward · · Score: 2, Informative

      The 160 GB SSD is probably 1-5x the size of your ipod...

    2. Re:Great. I buy a 160GB iPod and now they by Tarlus · · Score: 1

      Great. I buy a 160GB iPod and now they tell me I could have got a solid state one. They do?
      Where?
      --
      /* No Comment */
    3. Re:Great. I buy a 160GB iPod and now they by iamhassi · · Score: 5, Interesting

      "The 160 GB SSD is probably 1-5x the size of your ipod..."

      why do you say that? I can buy a 16gb flash drive for $60. Line 10 of those up and you have a 160gb flash drive for $600 that shouldn't be much bigger than a iPhone if you remove the unnecessary plastic and USB ports from the drives.

      Imagine a RAID0 array of ten 16gb flash drives! 200+ mByte/sec (ten x 20mB/sec) transfers and access times in nanoseconds vs hard drive milliseconds! No more bottlenecks.

      i for one welcome our new flash memory overlords!

      --
      my karma will be here long after I'm gone
    4. Re:Great. I buy a 160GB iPod and now they by __aaxwdb6741 · · Score: 4, Informative

      You forgot the 10x increased chance of unrecoverable failure.

    5. Re:Great. I buy a 160GB iPod and now they by SilverEyes · · Score: 1, Informative

      The average seek time for a hard disk is measured in milliseconds, but for continued transfers, they can have a much higher data throughput than a flash based device. File systems, caches, pages, compilers are all organized to take advantage of this ability to load lots of sequential data quickly, once a seek has been completed. It will probably take a few more before SSD make a huge difference in performance because of the use of existing technology. Now for small random accesses (i.e. one page), a SSD wins, hands-down.

      --
      Interesting.
    6. Re:Great. I buy a 160GB iPod and now they by Spokehedz · · Score: 1

      But considering that the MP3 player is all about random music playing, and the files are quite small, not to mention that it could greatly benefit from the reduced power consumption.

    7. Re:Great. I buy a 160GB iPod and now they by Culture20 · · Score: 1

      Okay, a RAID6 array, with a couple inline hot-swap chips.

    8. Re:Great. I buy a 160GB iPod and now they by ATMD · · Score: 4, Interesting

      Could that be why music players tend to have flash storage, while most PCs still have hard drives?

      --
      Nobody else has this sig.
    9. Re:Great. I buy a 160GB iPod and now they by jandrese · · Score: 4, Informative

      As far as random access on a drive is concerned, a 5MB music file is gigantic. The seek time (1 seek every 3-4 minutes) is a non-issue. If you were playing 20 snippits of different songs every second then it might matter, but for MP3 playing it is not an issue at all. Even if your file gets fragmented for some reason you're only going to be talking about a few dozen seeks at most.

      That said, flash does have a bunch of advantages for music players. It's far more shock resistant (for running!), requires less power, and doesn't have to constantly be put to sleep and woken up like spinning magnetic media.

      --

      I read the internet for the articles.
    10. Re:Great. I buy a 160GB iPod and now they by matt21811 · · Score: 1

      You can take solice in the fact that you saved some money over a flash based iPod of the same capacity. 160 gig of 1.8 inch disk is still a bit cheaper than 160 of flash storage. But not for long.
      I've been studying the price improvement of hard disks and flash for a few years now. If both of them keep improving at the same rate they have for the last 5 years then 1.8 inch drives wont be viable in 12 months from now. The flash will be cheaper.

      Graphs and data for this prediction here: http://www.mattscomputertrends.com/flashdiskcomparo.html

      In a way, you have bought a peice of computing history, the last of the disk based MP3 players.

    11. Re:Great. I buy a 160GB iPod and now they by iamhassi · · Score: 3, Insightful

      "The average seek time for a hard disk is measured in milliseconds, but for continued transfers, they can have a much higher data throughput than a flash based device."

      True, but PCs don't store data in consecutive order. Data is just placed haphazardly around, and it's up to the file allocation table to keep track of it all. So that 5 gigabyte game you're installing isn't all in one giant line of bits, it's shoved everywhere all over the drive, and it's constantly seeking to find where the rest of the files are to load the next level. That's why people periodically defragmenting their hard drives, to put the files all next to each other and save those precious milliseconds, which quickly turn into seconds when the PC's loading a ton of files into RAM.

      Because of fragmentation it's rare to have 60 megabytes of data for one application all next to each other, so that's why hard drives rarely read at there top speeds, they read a couple hundred kilobytes, seek 10ms, read some more, seek, etc.

      That's why people spend big $$$ to go from 7200rpm hard drives to 10k or 15k rpm SCSI drives, because just going from 8ms down to 3ms makes a very noticeable difference. So the jump from milliseconds down to nanoseconds would make a tremendous difference. RAM is measured in nanoseconds, so to have a 160gb drive only 5-10x slower than ram would be much better than the 1,000,000 times slower speed of hard drives accessing in milliseconds.

      --
      my karma will be here long after I'm gone
    12. Re:Great. I buy a 160GB iPod and now they by iamhassi · · Score: 1

      "but for continued transfers, they can have a much higher data throughput than a flash based device. "

      And that's why I said RAID-0 array, so the transfer speeds could be increased by spreading the files amongst all the flash drives.

      --
      my karma will be here long after I'm gone
    13. Re:Great. I buy a 160GB iPod and now they by Minozake · · Score: 1

      It could also be the problem with flash drives where they fail after so many writes.

      I've killed 3 already in the past 6 months since I was an idiot and I ran programs off of it that weren't made for flash drives.

      I'm kind of weary of a 160 GB flash drive where I didn't even get to use all of my 1 GB.

      --
      http://sourcemage.org/ - Have fun :)
    14. Re:Great. I buy a 160GB iPod and now they by mabhatter654 · · Score: 1

      flash chips are actually 9 or 10 bits wide, with the extra bits used for internal error correction... raid already included!

    15. Re:Great. I buy a 160GB iPod and now they by Zemran · · Score: 1

      Unfortunately the old market force of milk it to the max will kick in. People said that cathode ray tubes would be dead years ago because LCD screens are cheaper to make but the manufacturers still make lots of money from the current scheme so cathode ray tubes are still common. Even though solid state drives will continue to improve, hard disks will remain the norm for a long time.

      --
      I love stacking my barbecues in the shed at the end of summer - you can't beat a bit of grill on grill action.
    16. Re:Great. I buy a 160GB iPod and now they by torkus · · Score: 1

      I'd mod this flamebat at this point...if i had any points left.

      It's been established that you simply can not wear out a flash drive in a short period of time unless it has an underlying quality issue to begin with (or you're using ancient flash cards that do not have wear leveling). So maybe you had a batch of cheap/bad flash but you did not kill a flash drive in an average of 2 months use. If i remember the math, it's pretty much impossible to kill one in 6 months...you generally would need something like 1-3 years of sustained max-rate writing to the drive.

      --
      You can get rich if you own a politician, but you have to be rich to buy one in the first place.
    17. Re:Great. I buy a 160GB iPod and now they by torkus · · Score: 2, Insightful

      Good points, bad maths.

      Skipping that, the sustained transfer rate on SSD's has been going up A LOT recently. From SanDisk:

      SanDisk SSD SATA 5000 2.5" achieves a sustained read rate of 67-megabyte (MB)*/sec and a random read rate of over 7000 inputs/outputs per second (IOPS) for a 512-byte transfer3 ... SanDisk SSD achieves an average file access rate of 0.11 milliseconds

      Sustained read might be less than the top end desktop hard drives but the extremely low avg file access time you will see a VERY significant increase in performance in virtually all applications.

      And the best part about a SS:, is it's multiple parallel chips. There's a finite number of parallel data streams you can combine but it sill easily compensates for the lower individual data rate per chip. It's more a function of the controller chip and chip-to-chip wiring complexity. If you custom designed it, you could easily get a flash drive an order (or two) of magnitude faster in sustained read/write than a mechanical one.

      Keeping in mind that SSD's have been main-stream (though in the far upper tier) for what, about a year? I'm predicting (magic ball) that performance on SSD's will soon be able to greatly exceed classic hard drive technology. Mfgs will then use that advantage to offer other features that were impossible previously.

      Oh, and immagine if swap file wasn't a curse word?

      --
      You can get rich if you own a politician, but you have to be rich to buy one in the first place.
    18. Re:Great. I buy a 160GB iPod and now they by torkus · · Score: 1

      Why bother? As the technology grows they'll just do that internally - run more flash chips in parallen and you increate you throughput. SanDisk already is up to 67MB/sec sustained and 7000IPOS

      --
      You can get rich if you own a politician, but you have to be rich to buy one in the first place.
    19. Re:Great. I buy a 160GB iPod and now they by iamhassi · · Score: 1

      "And the best part about a SS:, is it's multiple parallel chips."

      that's RAID-0 on a chip. That explains the high read rate, although 67 megabyte/second isn't amazingly fast for flash drives: A single 1gb ATP Toughdrive reads at 33 mB/second. RAID-0 two of them and you have 66 mB/sec. Sounds like that's exactly what they did.

      Another benefit of flash memory over hard drives I forgot: no maximum vs minimum rates. Data can be placed anywhere and read/write at the same speeds, compared to hard drives where data stored on the outside of the platters reads and writes faster than the inner. portion.

      --
      my karma will be here long after I'm gone
    20. Re:Great. I buy a 160GB iPod and now they by Minozake · · Score: 1

      I did not realize this. Thank you for clarifying.

      --
      http://sourcemage.org/ - Have fun :)
    21. Re:Great. I buy a 160GB iPod and now they by WithLove · · Score: 1

      He's right. I can't cite any sources (in true Slashdot fashion!) but I remember reading in some tech journal that even if you completely formatted a flash drive into 0's and wrote all 1's onto it afterwards every hour, on the hour, it would still take a little over two years to wear it out.

    22. Re:Great. I buy a 160GB iPod and now they by Stephen+Samuel · · Score: 1
      That's 20000 writes. If you're using EXT[23] file system with access time updates enabled (the default), you can end up with 20000 updates to the superblock entries of popular files in pretty short order. Write levelling helps this, but it's not perfect. If you have a high number of writes to your drive, you're stil going to have problems a lot faster with flash than with many hard drive solutions.

      I saw a lot of queries, last year, on the EXT3-users mailing list from people who put ext[23] on an embedded flash application and found problems a lot sooner than they expected.

      Where Flash drives are in their glory is with read-mostly applications. MP3 players are a perfect example of this. You write a song once, and then play it back dozens or hundreds of times with no further modification. Computer applications where some blocks of data are continuously written to are a flash engineer's nightmare. Intel may have solved this problem, or they may just hope that the drive will last long enough to make the people who buy this solution happy.

      Remember that people who pay $4000 just for a hard drive are generally also the kind of people who will probably replace their machine every year or too, anyways. Those people will probably give away their flash-drive laptop before the drive is likely to fail.

      --
      Free Software: Like love, it grows best when given away.
    23. Re:Great. I buy a 160GB iPod and now they by Stephen+Samuel · · Score: 2, Interesting
      Music players tend to have flash drives because:
      • really low power consumption is rather important when your battery is the size of Bill Gate's Di...git.
      • It's only recently that flash drives have gotten big enough (4-20GB) for most people to seriously consider them as their primary computer storage device. 128KB, on the other hand is acceptable for storing your personal top-80 list (which is more than many music stations will play in a day.. modulo commercials).
      • It was only recently that hard drives got small enough to be able to fit in most MP3 players. Remember the ads about the guy with the 5 pound MP3 player? not too far from the truth if you use 5-year old HD solutions.
      • You write new songs {once a day ~ once a month}, then you listen to them {dozens ~ hundreds} of times. This read-mostly usage works just peachy with flash (which has to go through all sorts of contortions to be able to survive continual writes).
      --
      Free Software: Like love, it grows best when given away.
  3. Proof by Slashidiot · · Score: 3, Insightful

    More proof that competing companies are good for consumers. I just hope that toshiba and samsung have enough strength to come up with something that takes the lead from intel.

    --
    Tis women makes us love, Tis Love that makes us sad, Tis sadness makes us drink, And drinking makes us mad.
    1. Re:Proof by happy_smile · · Score: 1

      however, I think it still takes years until the price is affordable for me. Besides, I'm still concerning the limited write cycles it has.

    2. Re:Proof by Dionysus · · Score: 2, Informative

      Besides, I'm still concerning the limited write cycles it has.

      I'm not sure the limit on write cycles will be a major concern at those sizes, especially if you keep the drive maybe 50-75% full.
      --
      Je ne parle pas francais.
    3. Re:Proof by ILuvRamen · · Score: 1

      yeah, the more people competing, the lower the prices. And the #1 thing that SSDs need right now is a price drop! Instead of 1 or 2 companies releasing a massive bunch of flash chips all connected to each other practically as a joke or concept product for like $1000+ now hopefully they'll get serious and start developing something that can rival a normal hard drive's price. 160GB is fine for me when it comes to a system drive. I have a 250 and a 500GB storage drive for videos and files and all that but the system drive is where I want a fast reaction. My boot time is horrible cuz of all the hard drive IO.

      --
      Google's Super Secret Search Algorithm: SELECT @search_results FROM internet WHERE @search_results = 'good'
    4. Re:Proof by kestasjk · · Score: 1, Interesting

      It can be if you're keeping the OS on the drive. Parts of the system that are constantly written over, like the page file, can go through write cycles quickly. This is why you don't want to put an OS on a flash drive, but it's okay for storing normal chunky files.

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

      Ever heard of enhanced write filtering? When you make lots of frequent writes to flash, I would assume that companies that distribute or manufacture flash memory would definitely want to try REALLY hard disprove any negatives with the media. You don't have many problems with Linux in regards to this...but if you're running XP Embedded, you can trash flash memory as it constantly writes data out to the registry and the like.

      Read. http://msdn2.microsoft.com/en-us/library/ms838638.aspx

    6. Re:Proof by mmkkbb · · Score: 1

      Don't modern flash drives have controllers specifically designed to spread writes around the physical media?

      --
      -mkb
    7. Re:Proof by Facetious · · Score: 1

      I don't know about controllers, but I have heard some OSes are smart enough to do this. (No, I can't provide a link.)

      --
      Let us not become the evil that we deplore.
    8. Re:Proof by torkus · · Score: 1

      Wear leveling...

      Why does this need to be explained *every* time there's a thread about flash memory?

      Most flash drives can sustain max-speed writing for several YEARS before approaching their max.

      Put your OS and swap file on there (swap will LOVE fast flash) and enjoy.

      --
      You can get rich if you own a politician, but you have to be rich to buy one in the first place.
    9. Re:Proof by Facetious · · Score: 1

      Why does this need to be explained *every* time... Hello. This is Slashdot.

      Anyway, considering the number of magnetic drives that I've lost over the years, flash will be my medium of choice as the price continues to decline.
      --
      Let us not become the evil that we deplore.
  4. But can I afford them yet? by onefriedrice · · Score: 2, Insightful

    The price needs to drop a lot for me to consider one above the tried-and-true magnetic hard drive.

    --
    This author takes full ownership and responsibility for the unpopular opinions outlined above.
    1. Re:But can I afford them yet? by Pharmboy · · Score: 5, Interesting

      At the show in December, another article said:

      "In a short demonstration of an Intel solid-state drive at work in a laptop, Saleski showed that the drive could read and write 680MB of data and related storage in 24 seconds. The read and write speed of the solid state drive will be three to four times faster than that of most hard drives, and it will initially cost as much as three times as much as a hard drive, he said."

      If in a year they are twice the price of a regular hard drive, that is a bargain for some of us, if for no other reason that to use it as a swap drive for the OS and scratch drive for Photoshop. It would also making loading game levels much faster, so an 80gb version could make an affordable addition to a regular drive that has the OS.

      --
      Tequila: It's not just for breakfast anymore!
    2. Re:But can I afford them yet? by Anonymous Coward · · Score: 0

      Tried and true?

      There is no piece of hardware in my computer that pisses me off more than hard-drives. I say good riddance.

    3. Re:But can I afford them yet? by timeOday · · Score: 3, Insightful

      3x the sustained read/write at 3x the price of a winchester drive is too good to be true. Keep in mind the access time for SSD destroys a hard drive. When you consider the value of data on a drive, and what it costs to have a tech replace one, I'd think winchester drives will quickly be obsolete in PCs for business users.

    4. Re:But can I afford them yet? by somersault · · Score: 1

      Better hope that it doesn't drop too quickly into hard drive prices or you'll crash the heads

      --
      which is totally what she said
    5. Re:But can I afford them yet? by fuzzylollipop · · Score: 1

      Actually you are way off base. Current SSD drives are about HALF as fast as 5400 RPM drives in writing, and pretty close to even on reading, definitely not any faster. So you won't see any performance in loading games or using Photoshop.

    6. Re:But can I afford them yet? by justinlindh · · Score: 1

      You're right, but I think we'll be in dire need of some optimization in the scenarios you describe.

      I know that Photoshop allows you to specify a scratch drive, and I'm assuming there's a way to specify where to toss the swap file in Windows (I know this is simple in Linux by just creating the swap partition on the flash drive).

      The problems come in your game example: Yes, game level loading (and related assets) would be greatly improved, and the performance improvement is definitely welcomed. But you wouldn't ideally install the entire game on the flash drive. It's not uncommon for installed games to take upwards of 5 or 6 gigs. This, combined with swap/scratch/etc will fill space up very quickly (especially an 80gb drive, as you mention as a good companion drive). This space is a price premium, so you probably wouldn't want to install the entire game to this drive, at least, until the price came down to where it wasn't a price premium. I've yet to see a game that lets you install some assets in a different drive/partition than the base install, and I'd be surprised if game developers decide to take the time to implement this.

      So, I agree that it's a great idea in theory, but it's a bit more difficult to implement, unfortunately.

    7. Re:But can I afford them yet? by Chatterton · · Score: 1

      80GB is actually enough to install the system, the swap, some softwares and some games. I do it with a Raptor. Data are stored on 2 other 250GB drives. I have never filed my Raptor to render it unusable. Now I have only 6 games installed on my computer and when I am fed up of one, I uninstall it and load a newer one...

    8. Re:But can I afford them yet? by Tony+Shepps · · Score: 3, Funny

      I think the technologies have advanced enough that the modern hard drive should not be called a "winchester drive". The name actually referenced a specific model of IBM hard drive manufactured in 1973... with a capacity of 35 MB.

    9. Re:But can I afford them yet? by Detritus · · Score: 1
      The fundamental concept hasn't changed much, it's still a sealed flying-head disk drive. Other than scale, it looks very similar to modern drives.

      The original Winchester drive had two 30MB spindles, 30-30.

      http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3340.html

      Some observers have noted that the 3340 was known as "Winchester" because its development engineers called it a "30-30" (its two spindles each had a disk capacity of 30 megabytes), the common name of a rifle manufactured by the Winchester Company. Kenneth E. Haughton, who led the 3340 development effort, is reported to have said: "If it's a 30-30, then it must be a Winchester."

      --
      Mea navis aericumbens anguillis abundat
    10. Re:But can I afford them yet? by lagfest · · Score: 1

      But he wasn't saying that about current SSD drives, he was talking about the SSD drives that intel is releasing in Q2.

    11. Re:But can I afford them yet? by lagfest · · Score: 3, Interesting

      If you're on a workstation, plugging in a few 8 or 16 GB ram modules might be better than using a photoshop scratch disk.

      /not affiliated with metaram btw.

    12. Re:But can I afford them yet? by cecil_turtle · · Score: 2, Informative

      Current SSD drives are about HALF as fast as 5400 RPM drives in writing Really? Can you point me to a 5400 RPM drive that has a 90MBps sustained write speed? Because I'm pretty sure you can't. There are different speed SSD's, but the faster performing ones are easily on par with current spinning drives for transfer rate, and are WAY faster for random I/O. They are a noticeable improvement.
    13. Re:But can I afford them yet? by Arterion · · Score: 1

      Once you've loaded the game's data off the magnetic drive, I think a modern OS is going to try its best to keep that data first in RAM, then in your swap. Windows XP calls it prefetching, and Vista improves upon the technology, calling it SuperFetch. I'm sure there are similar technologies at use on linux and MacOS.

      So even if you don't install the whole game on your fast drive, you should still see benefits from having a large swap on it.

      If that wasn't good enough, you could install the game to a directory of your choosing on your magnetic drive, then use symbolic links pointing to the game's executables and resources of your choosing that you have copied to your fast drive. However, my experience with games would suggest that unless you moved ALL the resources, you're not helping yourself much. For example: if you move all the video resources, the sound resources will slow you down. In most games I've experienced, the performance of subsystems, such as audio, can affect the overall performance of the game. In some cases, disabling the audio gives a clear performance boost.

      Having said all that, I think using an x64 platform with a game compiled for it, and ample RAM (4+ GB), will help more than anything. If you avoid the need to fetch things off the disk in the first place (beyond the initial load), you're much better off. Even SSD's are woefully slow compared to RAM. And with today's RAM prices, installing 4GB or even 8GB is not terribly expensive. The unfortunate thing is that on Windows, a 32-bit process can only use up to 2GB of RAM, even on a 64-bit version of the OS.

      --
      "That which does not kill us makes us stranger." -Trevor Goodchild
    14. Re:But can I afford them yet? by jimicus · · Score: 1

      I tell you where that really wants to be.

      It wants to be in hotplug caddy with an SAS interface to go into servers.

      Disk I/O - and particularly seek time - can be a real killer on server performance, particularly with the current trend for virtualisation. And sysadmins and managers expect to pay 3-5x the price per GB compared to desktop storage anyway.

    15. Re:But can I afford them yet? by cafucu · · Score: 1

      yeah, and we probably shouldn't call our computers von neumann machines because it's too accurate a description, but sounds so old school.

      --
      :%s:work:/.:g
    16. Re:But can I afford them yet? by dfghjk · · Score: 1

      680MB in 24 seconds (28 1/3 MB/sec) is not 3-4 times faster than most hard drives, much less so in the future when such drives would become available. I would expect the new 500GB notebook drive to be able to do that.

    17. Re:But can I afford them yet? by mabhatter654 · · Score: 1

      and those cost 10x a HDD, not 3x. The macbook air is closest comparison of the drive in an actual shipping computer, and the numbers seem to be a wash. SDD is better at small, random files, but the HDD wins for the longer transfers. And the HDD they use is an iPod drive not a desktop drive. It's a good start, but needs about 2 more cycles to be comparable... that's about 18 months at this point... Intel looks to be at about the 9 month first turn (macbook air had to use existing drives to be produced, it's already 6 months out of date!)

    18. Re:But can I afford them yet? by bruce_the_loon · · Score: 1

      The fastest I've seen a SATA-2 to SATA-2 copy, different channels, for 700MB was 87 seconds. 680MB in 24 seconds translates to 230Mbps which is faster than any consumer drive I've seen around.

      And before someone 3.0Gbps's me, that's bus bandwidth, not drive. No drive can feed close to that.

      --
      Trying to become famous by taking photos. Visit my homepage please.
    19. Re:But can I afford them yet? by nosfucious · · Score: 1

      With 4+ Gig of ram, the first this I'll do is make a 1G ram disk.

      Then I'll put my swap file on it.

      Thats the best speed boost my swap file/partition ever got.

      --
      Q:I was listening to a CD in Grip and it sounded horrible! What's up? A:Perhaps you are listening to country music
    20. Re:But can I afford them yet? by Pharmboy · · Score: 1

      They pay 3x for SCSI drives mainly because of the testing they undergo, not the technology. I am not sure if the flashdrive systems are ready for server apps yet. That is one hellofa different kind of load than desktop service.

      --
      Tequila: It's not just for breakfast anymore!
    21. Re:But can I afford them yet? by torkus · · Score: 1

      Long file transfers aren't where your 'user experience' generally comes in. How often do you read more than 100MB sequentially off a hard drive?

      How often do you read 100MB in much smaller files (100KB - 5MB)? Ahem, booting windows, loading warcraft...and so on.

      So yes, HDD wins in the one area that means the least. I bet your car has the loudest muffler too.

      --
      You can get rich if you own a politician, but you have to be rich to buy one in the first place.
    22. Re:But can I afford them yet? by shutdown+-p+now · · Score: 1

      Curiously enough, the name "winchester" for hard drives has become a genericized trademark in Russian, and persists to this day, although usually in shortened forms. Could happen in other languages as well, and the GP might well be not a native English speaker.

  5. Bummer. by RandoX · · Score: 2, Funny

    I thought this was an announcement for a 160 gig USB thumb drive. Not that I could afford it anyway.

    1. Re:Bummer. by Joe+The+Dragon · · Score: 1

      at that size it better be a usb 3.0, firewire 400 or higher, or E-sata. Even firewire 400 is faster then usb 2.0.

    2. Re:Bummer. by Oktober+Sunset · · Score: 1

      why? Just cos I have 160 GB of storage doesn't mean I want to read/write the entire of it at once.

    3. Re:Bummer. by TeknoHog · · Score: 1

      why? Just cos I have 160 GB of storage doesn't mean I want to read/write the entire of it at once.

      You can't just write a set of data "at once". You're limited by Planck time, so the physics bottleneck of writing that 160 GB is about 2970000000000000000000000000000000000000000000000000000 bytes/s. Nothing to brag about.

      Kids these days.

      --
      Escher was the first MC and Giger invented the HR department.
  6. Logical move by crow · · Score: 5, Insightful

    It's very difficult to move into an established market, like disk drives. There's tons of technical expertise to acquire, and without your share of patents to negotiate a sharing deal, you're going to be paying through the nose in royalties. You just don't see new disk drive companies popping up. The only way to enter the market is to buy or partner with an existing player.

    The shift to flash drives changes all this.

    This is Intel's one chance to become a major player in a component that they haven't been involved in until now.

    1. Re:Logical move by msgmonkey · · Score: 1

      Intel have been in the flash memory market for as long as I can remember, they're just focusing on a growth market like everyone else is. AMD where also in the game, have n't really kept up with what they have been up to though.

    2. Re:Logical move by thrillseeker · · Score: 4, Informative

      I'm curious at what point we will quit treating these hard drive replacements as that, and instead treat them as what they are - large arrays of addressable memory. Without doing the homework to be sure, I suspect that being able to remove the overhead of an OS building the needed protocol stream to address this memory as a hard drive, and instead treating it as memory, would save significant(?) code/time.

    3. Re:Logical move by bendodge · · Score: 1

      Flash drives are very nice, but I want bigger SD cards (or something similar). That way I can buy a pocketful of cheap storage cards for something like the Eee. USB flash drives stick out too far. (Although the type without any housing on the plug somewhat cure that.)

      Is there some limiting problem for SD cards that prevents them from being 20GB or so?

      --
      The government can't save you.
    4. Re:Logical move by calebt3 · · Score: 1

      I don't think it'll happen before we get rid of the limited write cycles. At which point it would also be a true HDD replacement. Anyways, RAM is still faster (I think)

    5. Re:Logical move by TheRaven64 · · Score: 1

      There is very little overhead in the block device protocol (and they are still block devices - flash chips are typically not bit-addressable, they are collections of block that must each be erased in one go). Most of the overhead is in the filesystem and you still need that for flash.

      --
      I am TheRaven on Soylent News
    6. Re:Logical move by nuzak · · Score: 4, Informative

      Current hardware with simple wear leveling will give you about a solid year of continuous writes. That's writing 24/7, nonstop. I don't think a HDD could even survive that. For a consumer device, even under fairly heavy use, the hardware will be long obsolete before it runs out of write cycles.

      --
      Done with slashdot, done with nerds, getting a life.
    7. Re:Logical move by amorsen · · Score: 1

      Is there some limiting problem for SD cards that prevents them from being 20GB or so? Kingston has 16GB, is that close enough? The limiting problem is simply physical size. As flash cells shrink, more will fit. The latest jump just happened, so it'll probably be a year before we see the next doubling.
      --
      Finally! A year of moderation! Ready for 2019?
    8. Re:Logical move by morcheeba · · Score: 1

      Read the book "The Innovator's Dilemma" (summary of what I'm talking about here).

      The author argues that innovation in disk drives comes from new players. This was true of the transitions from 8" drives -> 5.25" -> 3.5" -> 2.5" -> 1.8" drives. Basically, older companies continue selling their products, oblivious to new markets (for which their older product doesn't fit). Eventually, the new market takes over and the old disk drive is useless. It took an ipod to make 1.8" mass-produced, and now I can buy a laptop that has one of these smaller drives in it. Desktops used to use 5.25" drives until luggables started making the 3.5" form factor popular.

    9. Re:Logical move by ODiV · · Score: 1

      That's very interesting, thanks for your comment.

      I find it intriging that the newer tech seems to have fewer licensing issues and barriers to market entry in this case. Normally, doesn't it go the other way? The older tech is accessable to all and the newer stuff is patented all to hell (eg: Anyone can make ground coffee, but if you make something too similar to my Coffee Pouch (TM) I have a case for patent infringement).

      Why is there no (obvious) such problems with flash memory? Why is it an open market? Is this a resurgence of an older idea?

      I know what I'm researching after work. :)

    10. Re:Logical move by Detritus · · Score: 1

      It's already been done. See the single-level store architecture of IBM's System/38, introduced in 1979. It was later replaced by IBM's AS/400.

      --
      Mea navis aericumbens anguillis abundat
    11. Re:Logical move by crow · · Score: 1

      My assertion that moving into a new market has fewer patent issues than moving into an established market is based on the speed of market maturity versus the lifetime of patents. Patents last for something like 18 years. That means that any hard drive technology from 1990 is open season for new companies to pick up. Obviously, to be competitive, you need to incorporate the more recent advancements, as 30MB RLL hard drives are useless today. Unlike the coffee maker market, patents here are effectively permanent--it's rare that a patent expires before the technology it covers is obsoleted by something else that is covered by more patents.

      So Intel can get into the storage industry now without needing any hard drive patents, except those that cover the SATA interface implementation (which they already deal with on the motherboard chipset side).

      As others have pointed out, Intel is already in the flash chip market, so stepping into the SDD drive market isn't really that big of a step technologically, but it is a big step from a business and marketing standpoint. Intel could have left the market to Seagate and the like, counting on selling them the flash chips to make their products work, but they're going for the whole pie.

    12. Re:Logical move by John+Whitley · · Score: 1

      I don't think it'll happen before we get rid of the limited write cycles Folks, get this through your heads: at the lowest level, NO mass storage media has acceptable reliability anymore. CD's have had error correction built in from the get-go. Magnetic hard drives have been using increasingly sophisticated layers of error correction, bad block marking, etc. for decades now. We will probably never again see a mass storage media that is high-reliability on the basis of a single storage unit (bit or whatever) again; the data densities are likely to just be too low to compete. We'll have a combination of strategies designed to work around the physical limitations of the media (e.g. write-leveling for flash) and layered redundancy/error correction strategies to deal with other failure modes.

      What's more: computational processes will probably enter the error-compensation era as well. There have already been articles floating across /. relating to work in this area.
    13. Re:Logical move by joe_bruin · · Score: 3, Informative

      No, this doesn't work. NOR flash can be addressed directly. This is not NOR flash, it's NAND. NAND can only be written and read in blocks. NAND requires block error correction due to the high incidence of bit errors (especially on multi-level cells). NAND uses complex wear-leveling software and a lot of black magic to work well (not just block remapping, but active block moving). NAND is very slow compared to real memory, and if you tried to read it as directly-mapped memory your processor would slow down to a crawl.

      NAND flash is really a block device. There's no getting around it. Some assumptions we make today will have to be thrown out (such as assuming there's an advantage to writing blocks close together or trying to reorder reads so the drive head sweeps in the most efficient manner), but in general access to NAND memory makes sense only through the same block serialization stack we use today for disks.

    14. Re:Logical move by mrtonic · · Score: 1

      I'd thought AMD had sold off their flash-based production... Yeah, it was previously a joint venture with Fujitsu and is now called Spansion.
      http://www.spansion.com/

  7. Could we see an end to Magnetic Media? by jellomizer · · Score: 3, Interesting

    Yes Megnetic Media is cheaper then Solid state... But higher speeds and still its prices are falling fast too, battery power usage, less points of failure. It really seems like the way to go. I could see Magnetic Media go the way of the CRT in 10 years? I think it is possible. Unless Magnetic makes some Huge Improvement in capasity and also we get a hug increase in demmand in data. Because drive size has began starting to exceed our data storage needs (at least on a personal computer Level)

    --
    If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    1. Re:Could we see an end to Magnetic Media? by brunes69 · · Score: 4, Insightful

      Because drive size has began starting to exceed our data storage needs (at least on a personal computer Level)

      Er.... I have several 30 GB HD rips that would tend to disagree with you.

      Multimedia content is still huge. Your standard from-the-factory PC can only hold 3-4 high quality movies. I know people who have multi-TB RAID arrays to archive their media content and are already feeling storage crunches.

    2. Re:Could we see an end to Magnetic Media? by calebt3 · · Score: 1

      But higher speeds and still its prices are falling fast too, battery power usage, less points of failure. But shorter lifespans. True, they are more predictable, but the average lifespan is still lower.
      I'm personally thinking of having a flash / drive and a HDD /home.
    3. Re:Could we see an end to Magnetic Media? by qortra · · Score: 1

      Hear Hear! Current storage capacity is just fine for people who use gmail and openoffice, but not for people who actually use their computers for media - especially video and uncompressed/losslessly compressed images and audio. I am one of those people who has a multi-TB linux server in the basement. I ripped my music collection to FLAC, and got about 1/3 of the way through my DVD collection (raw rips with menus), and now I need more space - significantly more space. If there is a be a brave new physical-media-less world, our current storage capacity is woefully unprepared.

    4. Re:Could we see an end to Magnetic Media? by AvitarX · · Score: 1

      Is that true?

      Because I don't think we have had SDD for long enough to really know what the average lifespan would be. Also it could very likely be increasing as HDD has been decreasing.

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
    5. Re:Could we see an end to Magnetic Media? by jellomizer · · Score: 2, Informative

      While there is a longer Lifespan for the data... Normally the machinancal parts of the drive die well before I have to worry about the magnitism going away. WHile Flash may have a shorter lifespan I would expect you will have more dependable drives due to the fact there is no moters to burn out or heads crashing down on the dive giving a nice scratch on it. And by the time it becomes a problem it is usally time for a major upgrade and you move the data over.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    6. Re:Could we see an end to Magnetic Media? by calebt3 · · Score: 1

      The data is still readable, right? Is it only the writing that fails?

    7. Re:Could we see an end to Magnetic Media? by asterix404 · · Score: 1

      This is not a factory PC drive, prices will have to come down a ton until the normal average person will ever buy one of these. On the multi-TB SAN however when you start getting into about 40-60HDD's with a 500k rack 16 fibre chans... all of the needed advances is because to get bits on disk is still way slower then bits being fed to them. Put solid state into a 12 fibre chan SAN at a back plain of 32 GB/s and you will see a performance boost from the slowest object in your SAN, which is in fact the most important place to see an improvement. That of course will only be used by people who will need that speed, IBM, UPS, and other mega tech houses will be the first to purchase them and I'm sure main frame DB techies are salivating at this possibility.

    8. Re:Could we see an end to Magnetic Media? by jellomizer · · Score: 2, Interesting

      Why do you think I gave it a 10 year span...

      I never stated that people will not be able to fill the drives but it is a case the demmand for space is less then the supply of space in general... Back in them olden days were drive size was in the 100s of Megabytes people were rather quickly filling up the drives with normal operations. This was true until drive size got over 40 Gigabytes. About 10 year ago... While drive size has increased we can still get by quite well with 40 Gig hard drive. Even with Vista Ultimate with Office 2003 and .NET 2008. This is the eqlivlant of in the year 2000 having windows 2000, Office 2000 and Visual Studio 6 running on a 512 Meg Drive... Our Demmand is actually less... Yes if you are going to some HD media (and most poeople don't... I don't... And I have a Mac too) It is really a special case of doing some unique work that normally you could afford to do with more... So Today except having a small array of 3 Terabyte Drives you will have a large array of 20 Solid State Drives. (expensive Yes) but doable and if the performance benefits help out it may be worth the cost... But as Flash Drive get cheaper and faster and larger I would predict that in 10 years The cost difference between Magnetic Drive and Flash Drives of the same size would be about $100 difference. ANd the gains would make it worth the extra cost.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    9. Re:Could we see an end to Magnetic Media? by Anpheus · · Score: 1

      The ratio between hard disk capacity and the size of the content we put on it is increasing, actually. Look at the trends, the launch date of the CD-ROM and DVD-ROM coincided with fantastically smaller hard disks than we have today, and they could hold proportionally less CD and DVD content without changing the compression. This is not idle speculation, this is cold hard fact.

      And this will only get worse. Already 1080P content requires you to be within 6 feet or so of a 100" screen to discern the difference between it and the next highest industry standard (2K resolution used in digital cinematography.) No one sits that close to a screen, they'd have to turn their neck just to see the whole thing. Our eyes simply don't have the resolution to keep increasing the definition, so HD is about as high as 2D pictures will go.

      No, I can't predict what will happen if holography becomes the norm, but I can tell you that the gap has historically widened between hard disk capacity and the ability to use that capacity. Hard disks have grown faster, in greater leaps than CPUs or bus speeds or even their own ability to read and write information. We have a wealth of information, and we'll increasingly find our tools inadequate to analyze it as we did before. That's unfortunate, but it's the way things are turning out. We'll have to come up with more and better ways to squeeze every byte and get every bit of value we can from our media because the only way we know of to improve our processing ability is to make the hard disks redundant: mirror the damn things until you have as much throughput as you need.

      Go back and look at standard from-the-factory PCs from the time of the creation of the CD-ROM and DVD-ROM. (Good luck with the CD-ROM one.) You'll find yourself lucky if you could store a CD or DVD rip or two on one of those things.

      P.S.: Try using a more compact encoding for your HD rips, you don't need all 30GB to have spectacular picture, H.264 can do amazing things.

    10. Re:Could we see an end to Magnetic Media? by DrVomact · · Score: 2, Interesting

      I could see Magnetic Media go the way of the CRT in 10 years? I think it is possible. Unless Magnetic makes some Huge Improvement in capasity and also we get a hug increase in demmand in data.

      Sure, it's possible. Ten years leaves you a bit of wiggle room. But I'm skeptical...I think hard drives will still be around in 20 years. Heck...I'll bet on 100 years—I won't be around to pay up.

      The reason I am skeptical of announcements of the impending doom of magnetic drives is that I first heard it in...let me see...1982 or thereabouts. People were talking about the "inherent limitations" of magnetic drives, but it wasn't clear at that time what could possibly replace them. Of course, what happened is that magnetic drive technology has proven astoundingly resilient: we went from huge platters inside drive bays the size of a washing machine that collectively held maybe 2M to cheap standard-sized drives that hold half a terabyte (street price currently hovering around 100 inflated US dollars).

      Don't get me wrong: I'm excited about the idea of replacing the boot drive on my PC with a super-fast flash drive—once the price for an 80G flash-based drive gets down to under $200. However, I think magnetic drives will maintain their huge price-per-MB of storage space advantage over rival technologies for a long time to come. So the drives in my PC that hold my video and movies will still be magnetic. I just want fast boot times, and quick swapping. Notebooks are another story, of course—I think it's quite likely that most notebooks will no longer have magnetic drives in ten years.

      But for solid-state storage devices to make magnetic media drives completely obsolete, two things would have to happen:

      1. Magnetic drive technology would have to hit a capacity "wall"—a point at which it becomes more expensive to make a standard-size (fits in a PC bay) drive of X capacity than an equivalent solid state device.
      2. Manufacturing costs for solid state devices having X capacity can be brought down to the point where it's profitable to sell the product at a price consumers will pay.

      These two points are related of course; they boil down to saying that it's going to have to become darn cheap to make a huge solid state "drive", where "huge" will probably be defined in tens of terabytes.

      --
      Great men are almost always bad men--Lord Acton's Corollary
    11. Re:Could we see an end to Magnetic Media? by ProfessionalCookie · · Score: 1

      Survey says 5 years, tops.

    12. Re:Could we see an end to Magnetic Media? by ThousandStars · · Score: 1
      Sure it could, over the late medium- to long-term. But my iMac has a 500 GB HD in it while a 64 GB SSD would represent 2/3 of the iMac's price. Even with Intel's announcement, we're not going to see any price/GB ratios that will make heavy inroads through desktops for a long time.

      Laptops... maybe, but we're still a few years away from seeing this approach the mainstream.

    13. Re:Could we see an end to Magnetic Media? by fishbowl · · Score: 1

      I'm still waiting to hear from one person who has had a flash device fail due to the limited number of write cycles.
      This is the excuse for not using flash memory in certain applications. But it's cheap enough to make it a part that
      is routinely replaced, for instance, and it could be made so that the duty cycle is predictable (unlike consumer flash
      devices that go to great lengths to hide the issue from the user.)

      --
      -fb Everything not expressly forbidden is now mandatory.
    14. Re:Could we see an end to Magnetic Media? by nine-times · · Score: 1

      I'm not trying to be antagonistic here, but I don't think your needs are exactly "the norm". Lots of people use their PCs for e-mail, word processing, web browsing, and MP3s. For lots of people, you won't see them ripping and storing HD movies until they have a set-top box that lets them watch it on their TV, and though that idea is starting to play out, it's still not clear what form that will take. It may be that we all have a specialty piece of hardware for HD DVR, IPTV, VoD, etc. in a few years, and then we still won't need high-capacity storage on our PCs, because it'll all be in this specialized set-top box.

      I think you're assuming two things that aren't necessarily true:

      1. Everyone will be ripping movies soon the way people have been ripping music: The problem here is primarily that people watch movies differently than how they listen to music. I can wear tiny headphones and get a decent music-listening experience from a portable player. I can't really get a decent movie-watching experience form a portable player or small computer screen. Also, by the masses have caught on to the phenomenon of hooking a computer to their TV for the sake of watching movies, it may be that physical media is out and streaming is the default.
      2. People care about quality and will require perfect-quality HD movies: People don't care as much about quality as you might think. Most people don't rip their music collection higher than 128kbps, let alone FLAC. Loads of people have spent money for a huge HDTV, plug it into a SDTV signal, and still marvel at how great it looks.

      I think it's very possible that purchasing videos will lose a bit of ground in favor of VoD to a set-top box, and then purchasing movies will be an over-the-internet media-less process that goes straight to your set-top box without going to your PC. In that's the case, your PC still won't need much storage.

    15. Re:Could we see an end to Magnetic Media? by brunes69 · · Score: 1

      P.S.: Try using a more compact encoding for your HD rips, you don't need all 30GB to have spectacular picture, H.264 can do amazing things.

      Sorry - I like to have my media archived in it's original format. Re-encoding already compressed data is just a recipie for lossiness, which is not what you want in a permanant archive.

      If I had the original uncompressed material to do a H.264 encode on, that would be an option, but when you only have the MPEG2 HD copy, it isn't.

    16. Re:Could we see an end to Magnetic Media? by 1310nm · · Score: 1

      I bet they have plenty of time to watch/listen to all that media while putting together said arrays, too. Always funny how media "collectors" probably couldn't cite half of the stuff they have stored.

    17. Re:Could we see an end to Magnetic Media? by TooMuchToDo · · Score: 1

      I have several Netgear 1U NAS devices in a 19" rack at my home, totaling almost 17TB of usable storage. I'm still running out of space.

    18. Re:Could we see an end to Magnetic Media? by dfghjk · · Score: 1

      Exactly. Capacity needs have not been surpassed by current products ESPECIALLY in notebook drives where these flash devices are targeted. The prediction of 10 years isn't very bold though. A whole lot can happen in 10 years.

    19. Re:Could we see an end to Magnetic Media? by Jarik_Tentsu · · Score: 1

      Agreed.

      If you just 'use' your computer at its basics - install a couple of games, have music, and some documents, you will never use up even 160GB. Most of my heavy gamer friends don't use that much either.

      Of course, when you get into media...I have 1.7TB, and it's all full. My anime spans 600GB, with the rest of my...*cough* movies and TV shows *cough*...taking up a large portion of the rest. My music is around 50GB. The drive for temporary video editing is usually around 300GB full at any one time (for obvious reasons)...

      And then, my C: System partition is a total of 40GB and is rarely even half full.

      If you store a lot of (video) media you're gonna need shitloads of space, if you're the kinda guy to 'watch and delete', 'watch and burn' or not watch at all, then in comparison, you'll need *nothing*.

      It's interesting how the market is split like that...with people like me looking at 2TB and my friends probably never needing more than 160GB.

      `Jarik

  8. Partition Filesystems by calebt3 · · Score: 5, Interesting

    What filesystem (NTFS, ext3, etc) is best for solid-state drives anyways? All of our commom filesystems are written for spinning drives, and certain features (such as ext3 self-defragmentation) probably shorten a flash drives lifespan.

    1. Re:Partition Filesystems by peragrin · · Score: 1

      Can someone with more knowledge,tell me if ZFS is better for flash drives over say NTFS or HFS+.

      I am wondering if OS X's slow move to ZFS has some unexpected side effects.

      --
      i thought once I was found, but it was only a dream.
    2. Re:Partition Filesystems by von_rick · · Score: 4, Insightful

      Why would you need defragmentation when there's no read head to consider? The whole idea behind defragmenting programs is to gather a file at one place so that the head doesn't have to jump to different addresses on the cylinder.

      --

      Face your daemons!

    3. Re:Partition Filesystems by calebt3 · · Score: 4, Interesting

      Exactly my point. Ext3 defragments itself automatically, which does more harm than good on a flash drive.

    4. Re:Partition Filesystems by Abcd1234 · · Score: 1

      probably shorten a flash drives lifespan.

      Well, that's probably only true on drives that don't automatically do wear leveling. Is it a waste of effort? Sure. But the additional write cycles will probably have a negligible effect given the sheer number of memory cells available in a device such as this.

    5. Re:Partition Filesystems by vmalloc_ · · Score: 1

      Most filesystems optimize for spinning disks primarily by starting the layout of the partition on the inner part of the spindle. The reason they do this is that it's much faster for the head to read these sectors than the ones on the outer part of the drive. So when you do this, the Operating System gets the fastest access part of the hard drive, which makes sense since your computer frequently accesses this to keep your system running.

      There's probably a ton of other optimizations like this, you are on to a very good point. A good person to ask this question would be Kirk McKusick, who is the genius behind FFS and a lot of other great filesystem ideas.

      -Kyle

    6. Re:Partition Filesystems by Telvin_3d · · Score: 1

      Even if it has a minor effect, why should we be willing to put up with any degradation at all?

    7. Re:Partition Filesystems by mapkinase · · Score: 1

      So what is flash drive lifespan nowadays in terms of read/write ops?

      --
      I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
    8. Re:Partition Filesystems by Aaron+Isotton · · Score: 1

      I'm not sure, but is it not the outer part which is faster (because it's moving faster past the head)? I don't see how seek time should be different on the inner/outer cylinders.

    9. Re:Partition Filesystems by AaronW · · Score: 3, Informative

      I was reading up on this a while back and it was recommended to use EXT2 instead of EXT3 since the journal would cause a lot more wear on the flash.

      I think there is definitely room for a Linux filesystem that is optimized for dealing with flash devices and limits the number of times data must be written. Furthermore, don't pad with 0's but with 1's (erased flash has all the bits as 1's).

      I would love to see a simple universal flash filesystem which could be used by portable devices and PCs without all the limitations of FAT32 (i.e. 4GB file limit) which seems to be the current fs of choice for consumer devices.

      JFFS2 is not suitable for regular flash drives (SD/MMC/CF/etc.) since it has its own wear leveling support and is optimized for devices without hardware wear leveling.

      For non-flash devices I have switched to XFS due to the higher performance and better tools compared to EXT3.

      -Aaron

      --
      This post is encrypted twice with ROT-13. Documenting or attempting to crack this encryption is illegal.
    10. Re:Partition Filesystems by Jeff+DeMaagd · · Score: 1

      I think wear leveling probably potentially has significant limits that proponents seem to ignore. Especially if you have less than 10% free space. Then all your writes will "wear level" over the available free space and degrade orders of magnitude faster than the maker suggests.

      That said, I haven't found a decently detailed write-up on exactly how wear-leveling accomplishes its task. I'm assuming that stuff that is rarely written will end up occupying space that's not available to the wear leveling algorithm until it is modified.

    11. Re:Partition Filesystems by calebt3 · · Score: 1

      Ahh... Wikipedia has some data on Flash-oriented filesystems: http://en.wikipedia.org/wiki/YAFFS http://en.wikipedia.org/wiki/JFFS2

    12. Re:Partition Filesystems by nuzak · · Score: 1

      > Why would you need defragmentation when there's no read head to consider?

      Because contiguous reads and writes are still faster than scattered ones. This means you have to avoid small fragments anyway -- once the fragments are big enough, making them all adjacent won't help much.

      --
      Done with slashdot, done with nerds, getting a life.
    13. Re:Partition Filesystems by calebt3 · · Score: 4, Informative

      Hmm...
      I just found this:

      Unlike DRAM, flash memory chips have a limited lifespan. Further, different flash chips have a different number of write cycles before errors start to occur. Flash chips with 300,000 write cycles are common, and currently the best flash chips are rated at 1,000,000 write cycles per block (with 8,000 blocks per chip). Now, just because a flash chip has a given write cycle rating, it doesn't mean that the chip will self-destruct as soon as that threshold is reached. It means that a flash chip with a 1 million Erase/Write endurance threshold limit will have only 0.02 percent of the sample population turn into a bad block when the write threshold is reached for that block. The better flash SSD manufacturers have two ways to increase the longevity of the drives: First, a "balancing" algorithm is used. This monitors how many times each disk block has been written. This will greatly extend the life of the drive. The better manufacturers have "wear-leveling" algorithms that balance the data intelligently, avoiding both exacerbating the wearing of the blocks and "thrashing" of the disk: When a given block has been written above a certain percentage threshold, the SSD will (in the background, avoiding performance decreases) swap the data in that block with the data in a block that has exhibited a "read-only-like" characteristic. Second, should bad blocks occur, they are mapped out as they would be on a rotating disk. With usage patterns of writing gigabytes per day, each flash-based SSD should last hundreds of years, depending on capacity. If it has a DRAM cache, it'll last even longer.

    14. Re:Partition Filesystems by asm2750 · · Score: 1

      Its suggested you use a FAT partition for max life for flash based drives because it doesn't talk that much.
      If you use a talkative file system like NTFS or ext3 I've some articles that state 5 year life is typical for flash rives.
       
        Honestly I don't think flash will be the end all solution that will make us use solid state drives, really its just a stepping stone to drives that utilize phase-change or another type of NV memory that has no real write fatigue and higher density.

    15. Re:Partition Filesystems by Sterling+Christensen · · Score: 1

      I've heard that modern implementations will move existing data around to keep the free space fresh enough for wear leveling.

    16. Re:Partition Filesystems by GodsMadClown · · Score: 1
      Flash file systems

      ( from http://en.wikipedia.org/wiki/File_system#Flash_file_systems; emphasis is mine)

      A flash file system is a file system designed for storing files on flash memory devices. These are becoming more prevalent as the number of mobile devices is increasing, and the capacity of flash memories catches up with hard drives.

      While a block device layer can emulate a disk drive so that a disk file system can be used on a flash device, this is suboptimal for several reasons:

              * Erasing blocks: Flash memory blocks have to be explicitly erased before they can be written to. The time taken to erase blocks can be significant, thus it is beneficial to erase unused blocks while the device is idle.
              * Random access: Disk file systems are optimized to avoid disk seeks whenever possible, due to the high cost of seeking. Flash memory devices impose no seek latency.
              * Wear levelling: Flash memory devices tend to wear out when a single block is repeatedly overwritten; flash file systems are designed to spread out writes evenly.

      Log-structured file systems have all the desirable properties for a flash file system. Such file systems include JFFS2 and YAFFS.
    17. Re:Partition Filesystems by ghjm · · Score: 1

      No, really, they aren't. If you just read block 2000 from flash media, a subsequent read of block 2001 and a subsequent read of block 546725 execute in exactly the same amount of time. Most modern flash devices randomize the actual locations to distribute wear evenly, so when you request logical block 2000 you might be getting physical block 71541, and then when you request logical block 2001 you might get physical block 391515. This makes a mockery of the very concept of defragmentation. You would be spending huge amounts of time and consuming the limited write lifespan of the flash device, only so to make sure that any file containing block 71541 stores its next piece of data on block 391515. It's meaningless.

      The OP is quite correct that once flash-based SSDs become more commonplace, all the major filesystems are going to have to adapt to perform optimally in that environment. Anyone looking forward to ext4 and NTFS 3.2?

      -Graham

    18. Re:Partition Filesystems by Abcd1234 · · Score: 1

      Because it's not worth the trouble to find a "better" filesystem if you've got one that already works reliably and well?

    19. Re:Partition Filesystems by warmflatsprite · · Score: 3, Informative

      JFFS2 is developed specifically for embedded devices, but I think that is because at its time of development the expectation was that only embedded devices would use flash media for primary storage.

      It's been a while since I've looked into how it works, but I'm speculating that it attempts to spread out write operations over the entire disk by giving file fragments fairly dynamic addresses. I believe it also has an ECC scheme and uses a reserved storage area for marking bad blocks. Since the SSD almost definitely does the last two in its controller hardware, JFFS2 would be a good option if you can make use of its distributed write operations only.

    20. Re:Partition Filesystems by Spoke · · Score: 4, Informative

      Ext3 defragments itself automatically No it doesn't. While ext3 does try to keep files contiguous and inodes in directories close to one another, it definitely does NOT do any defragmentation. ext2/3 filesystems have a history of getting highly fragmented over time and it gets worse the less free space you have on the disk.

      The ONLY way you can defragment a file is to copy the fragmented file to another partition, remove it and copy it back. If you want to defragment a complete ext2/3 filesystem, make a backup of the filesystem using tar, delete the original and restore the backup.

      No, this is not something you want to do while other software may be looking for the file.

      Of the common filesystems available for Linux (ext2/3, xfs, jfs, reiserfs) the only one that supports online defragmentation is xfs (using the xfs_fsr utility) and this has to be scheduled manually.

      Fragmentation in ext2/3 files is a huge problem when appending to files over long periods of time. You can check the fragmentation of any file on ext2/3 using the filefrag utility. Make a copy of a highly fragmented file (even to the same partition) and you will see the number of fragments go down dramatically, unless you don't have much free space left on the partition and the space you have free is also highly fragmented.
    21. Re:Partition Filesystems by Abcd1234 · · Score: 2, Informative

      I think wear leveling probably potentially has significant limits that proponents seem to ignore.

      Well, I think the earth probably is the center of the universe. 'course, both of our statements are unsupported by anything but guesswork, so why should either statement be believed over the people actually working in the industry on wear-leveling technology in modern flash drives?

      Especially if you have less than 10% free space.

      a) 10% of a 160GB flash drive is still 16GB... plenty of space, even if you are concerned.
      b) Most people don't run their hard disks anywhere near 10% of their storage capacity, so it's a minor issue.
      c) Wear leveling drives may very well contain addition free storage space just so that the algorithm can perform optimally.
      d) Using a log-structured filesystem, it doesn't really matter how much free space there is, as the free space region always moves through the disk... hence the "leveling" part of wear leveling.

      That said, I haven't found a decently detailed write-up on exactly how wear-leveling accomplishes its task

      Well, you could just start with Wikipedia, rather than blindly speculating.

    22. Re:Partition Filesystems by amorsen · · Score: 2, Interesting

      No, really, they aren't. If you just read block 2000 from flash media, a subsequent read of block 2001 and a subsequent read of block 546725 execute in exactly the same amount of time. In the beginning, back in the days of interleave, hard drives were pretty close to random access. Tape drives had around the ratio of transfer speed to seek speed that hard drives have today. At one time RAM was truly random access as well, now reading the next byte is often more than 10 times faster than reading a random one. The same thing is happening to flash. Of course it will be decades before the problem will be as big as the one we have with hard drives now, but it will happen.
      --
      Finally! A year of moderation! Ready for 2019?
    23. Re:Partition Filesystems by hoggoth · · Score: 0, Troll

      > The ONLY way you can defragment a file is to copy the fragmented file to another partition, remove it and copy it back. If you want to defragment a complete ext2/3 filesystem, make a backup of the filesystem using tar, delete the original and restore the backup.

      I call bullshit.
      Free up a lot of space, then copy every file over. Something like this:

      find / -exec (cp {} {}.defrag; rm {}; mv {}.defrag {});

      Done.

      --
      - For the complete works of Shakespeare: cat /dev/random (may take some time)
    24. Re:Partition Filesystems by afidel · · Score: 1

      Actually I can see where scattered reads are FASTER than contiguous reads, if you can read from multiple physical chips in parallel that would probably be faster than reading from adjacent cells in one chip.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    25. Re:Partition Filesystems by Spoke · · Score: 4, Informative

      I call bullshit. Why is it that when someone things someone else is wrong and they reply, they must make some smart-ass comment like "I call BS?" or "You suck, I'm right you're wrong?". Come on, keep it civil. Or should I say "I call bullshit".

      find / -exec (cp {} {}.defrag; rm {}; mv {}.defrag {}); Anyway, your defragmentation method is exactly the method I described. It won't defrag as well as backing it up completely, deleting all the original data and restoring because you are limited by the amount of (fragmented) free space you have. If your partition is close to empty, then it won't make much of a difference.

      However, your defrag method IS NOT SAFE and WILL RESULT IN DATA LOSS on a live system (sorry to yell, but I don't want anyone trying it on a live system - it should be OK if you can guarantee that no-one else will modify the data on the partition)

      There is a lot of opportunity in your script for data loss:

      1. During the copy. If someone modifies part of the original that has already been copied, your .defrag file won't have those changes.
      2. During the rm. Deleting files takes time, so there is more room for a file to try to write to the original. This step is actually completely unnecessary, just overwrite the original with your mv command.
      3. After the mv. If a process has the original file open, it will continue writing to that original file, even after it's been deleted and "overwritten". It is very legal to continue file operations on an open file descriptor.

      I suggest you actually try your defragmentation trick on a live filesystem which is actively in use. If you don't lose data, you're lucky.

      So I'll say it again. The only filesystem which allows you do perform LIVE defragmentation is xfs using it's xfs_fsr utility.
    26. Re:Partition Filesystems by DamonHD · · Score: 1

      You must be new round here! B^>

      Rgds

      Damon

      --
      http://m.earth.org.uk/
    27. Re:Partition Filesystems by Anonymous Coward · · Score: 0

      The self defragmentation probably shortens the lifespan of your mechanical drive just as much.

      I imagine when more people are using solid-state drives the old filesystems will just be tweaked to account for it. At least until some equivalent new filesystem designed specifically for solid-state devices comes around. There already are such filesystems but none have all the features of a modern filesystem like ext3 or XFS. I also imagine a good solid-state filesystem would be significantly simpler than anything currently used.

    28. Re:Partition Filesystems by hoggoth · · Score: 1

      ok ok...

      So how about I wrap it in a smarter script that uses 'lsof' to check if a file has an open file descriptor before overwriting it. I skip open files, or make a list of them and try them again at the end of the run. That should do it. Or I use something like 'snapfs' to shapshot changing files.... I guess I'm moving far from plain ext2/ext3 now...

      --
      - For the complete works of Shakespeare: cat /dev/random (may take some time)
    29. Re:Partition Filesystems by Spoke · · Score: 1

      If it were simple, someone would have written an online ext2/3 defragmenter a long time ago. There have been plenty of people asking for it, just google it.

      There is an offline ext2 defragmenter, but it hasn't been maintained in ages from what I understand - I don't think it's trusted to defragment your data without risking data loss.

      Both of your lsof and snapfs suggestions leave room for data loss unless you are sure that no one will be modifying the files. For that case, there is already a tool which attempts to defrag files called shake.

    30. Re:Partition Filesystems by Anonymous Coward · · Score: 0

      This makes a mockery of the very concept of defragmentation.
      I'm not sure that's necessarily true. Remember that a SSD is composed of multiple flash modules. It takes longer to read 4 blocks from one flash module than it does to read 4 blocks from 4 different flash modules. So having logical blocks could still be useful since it allows the wear-leveling logic in the drive to optimize performance based on its internal workings.

      Getting back to fragmentation, a fragmented file could end up with a disproportionate number of bits stored on one flash module. If that happened, reading that file from disk would be slower than the defragmented equivalent. The odd part is, the defragmented file would actually end up being strategically fragmented on the physical storage media, so we'd have to get used to the notion of physical fragmentation vs. logical fragmentation. In traditional HDDs, the two are one and the same. In flash-based SSDs, the two are very different.
    31. Re:Partition Filesystems by hoggoth · · Score: 1

      > Both of your lsof and snapfs suggestions leave room for data loss unless you are sure that no one will be modifying the files

      No, snapfs guarantees all writes will be sent to the snap filesystem, not the main file.
      If the snapfs creates a modified version of the file, the defragger should not overwrite the original, but instead at the end of the run copy the snapfs versions of all modified files back. No data loss can occur.
      That's what snapshots are for.

      --
      - For the complete works of Shakespeare: cat /dev/random (may take some time)
    32. Re:Partition Filesystems by Spoke · · Score: 1

      No, snapfs guarantees all writes will be sent to the snap filesystem, not the main file. But snapfs hasn't been maintained for 8 years. Does it even work? It says right on the homepage it is "incomplete, abandoned and unsupported" and we are talking about ways to defrag your filesystem TODAY.

      Even the basic "make a copy of the file" defrag methods we've been discussing aren't even guaranteed to leave the file you are attempting to defrag in fewer fragments than the original. In a lot of my testing, often the "make a copy of the file" defrag method doesn't always help for files only moderately fragmented on an old and fragmented filesystem.

      So I'll say it again: If you want to defragment files on Linux, basically your only real-world capable choice is XFS using the xfs_fsr tool.
    33. Re:Partition Filesystems by Blakey+Rat · · Score: 1

      The ONLY way you can defragment a file is to copy the fragmented file to another partition, remove it and copy it back.

      I call bullshit.
      Free up a lot of space, then copy every file over. Something like this:

      find / -exec (cp {} {}.defrag; rm {}; mv {}.defrag {});

      Done.


      How is your solution any different than the original example? Except yours assumes the free space is on the same partition, and the original assumes it's on a different partition.

      In any case, yours isn't a "live" defrag, unless you're OK with losing vast amounts of data, so I don't see why you're "calling bullshit."

    34. Re:Partition Filesystems by klui · · Score: 1

      Based on the Wikipedia entry below your response, it appears wear leveling will take into account the whole disk and will swap entries regardless if they're "in-use" or not.

    35. Re:Partition Filesystems by hoggoth · · Score: 1

      > I don't see why you're "calling bullshit."

      Ummm... because I'm being unreasonably antagonistic to provoke a response?

      --
      - For the complete works of Shakespeare: cat /dev/random (may take some time)
  9. Re:I for one... by calebt3 · · Score: 3, Funny

    Why? A solid-state overlord is not much more than a geometric rock.

  10. I'm curious... by Oxy+the+moron · · Score: 0

    What niche of the computer world will these disks fill at their current price point?

    First, you'd have to know who would want to buy them. Enthusiasts? I suppose... but I think most enthusiasts want the best bang/buck ratio for storage space, and why buy one of these when you can get a good ol' magnetic drive with a capacity of 1TB? People who require "rugged" notebooks? Solid state is definitely a plus here, but the main organizations I know that use them are mostly government (public safety, et al) and I don't see them affording it. What about servers? Obviously 160GB is overkill for a router-type box, and the smaller 4-8GB solid state drives would be a much better fit. Perhaps the price point would be right for a server that only serves web pages or simple file storage? Again, I think here that most people have some form of large magnetic storage (perhaps via SAN) and a web server would just pull from that.

    This kind of thing is really slick... I just don't see it being accepted in the near future.

    --

    Proudly supporting the Libertarian Party.

    1. Re:I'm curious... by imamac · · Score: 1

      This kind of thing is really slick... I just don't see it being accepted in the near future. I think the same was said of flat-panel computer monitors. Now that's pretty much all you can find.
    2. Re:I'm curious... by Oxy+the+moron · · Score: 1

      I think the same was said of flat-panel computer monitors. Now that's pretty much all you can find.

      Well... yes... but it took a while for flat panels to catch on. This is my point now, as well. I think it will *eventually* be widely accepted... just not in the near future.

      --

      Proudly supporting the Libertarian Party.

    3. Re:I'm curious... by 74nova · · Score: 1

      business laptops like mine. I'm not sure I could convince the boss to let me buy one yet, but I run virtual machines on it for testing and do a fair bit of other development. If I were to have a very fast drive in it, swapping for all the virtual machines would be faster and I could give them less memory. I don't know, it all depends on what these things cost.

      --
      use your turn signal! you people act like it's divulging information to the enemy
    4. Re:I'm curious... by imamac · · Score: 1

      I can see that, depending on your individual definition of "near" as it relates to technology. I would bet in less than 5 years most all laptops will ship with these. I suppose that's "a long time" in the computer technology world, though.

    5. Re:I'm curious... by calebt3 · · Score: 1

      Define "near". Next week? Next couple of months? Right after Duke Nukem Forever is released?

    6. Re:I'm curious... by von_rick · · Score: 1

      If the Magnetic Vortex Core (http://www.mpg.de/english/illustrationsDocumentation/documentation/pressReleases/2006/pressRelease200611281) technology ever makes it into the hard-drive markets, it will considerably reduce the size of these drives. Densely packed cores would mean less movement, thus lesser power, and higher stability - and yes, less weight. SSDs would be quite expensive and slow when compared with drives built on Vortex Cores. Lets see what the future has in store.

      --

      Face your daemons!

    7. Re:I'm curious... by nuzak · · Score: 2, Funny

      Plus, anything called a "Magnetic Vortex Core Drive" is a damn cool piece of hardware to own.

      --
      Done with slashdot, done with nerds, getting a life.
    8. Re:I'm curious... by businessnerd · · Score: 1

      There's already plenty of demand. As you said, the laptop market is one of them, but not just the "rugged" laptop market. The "thin" laptop market is one of them. The MacBook Air has a solid state drive (optional?) and the Lenovo ThinkPad competitor has one standard. Any laptop would benefit greatly from an SSD. For laptops, the HD is the last moving part, with the exception of cooling fans. Reliability is a key benefit to these drives (when used with a proper file system) because they don't have moving parts that can break down. Hard drive failure is the #1 hardware problem for ANY type of computer. Just ask any support desk. If spending a little more up front on SSD means not having to replace the drive down the road, then I'm sure plenty of organizations will want to make the switch. Even on servers, there is demand. Why do we have all of these SAN's with RAID 5 and 10 arrays? Because hard drives fail so often. We might not be getting rid of the RAID, but we will have less downtime and be replacing less drives.

      --
      "It's not whether you win or lose, it's how drunk you get." -- H. J. Simpson
    9. Re:I'm curious... by couchslug · · Score: 1

      "What niche of the computer world will these disks fill at their current price point?"

      Robust portable storage. Put one in a rugged USB enclosure and they would be dandy for carrying all my stuff and booting my OS of choice on different computers.

      --
      "This post is an artistic work of fiction and falsehood. Only a fool would take anything posted here as fact."
    10. Re:I'm curious... by Anonymous Coward · · Score: 0

      >Well... yes... but it took a while for flat panels to catch on

      From my perspective it was one of the fastest paradigm shifts ever. Shockingly fast, even.

    11. Re:I'm curious... by ODiV · · Score: 1

      From my perspective it was one of the fastest paradigm shifts ever. Shockingly fast, even.

      I'm with you there. Though there was a bit of a lag between a select few users having them and regular consumers buying them, there was definitely very little time between some consumers buying them and "everyone" having one.

      It ranks just behind DVD players in my (flawed?) memory as far as mass adoption goes. It only takes one government year end paired with a holiday season with new devices at "reasonable prices".

  11. I'm an idiot by dj245 · · Score: 2, Interesting

    But shouldn't these figures be some more convenient power of 2? Like 64GB (rounded) or 128GB?

    --
    Even those who arrange and design shrubberies are under considerable economic stress at this period in history.
    1. Re:I'm an idiot by slashgrim · · Score: 4, Informative

      160 = 10 x 2^4. So, probably 10 x 16GB chips

    2. Re:I'm an idiot by crow · · Score: 1

      As the other poster mentioned, these are hard drive GB, not memory GB, so it's base 10, not 2 when it comes to advertising the numbers. Also, I'm not sure how the wear leveling and failure handing works, but they may have some small extra space set aside to replace blocks that wear out, just like you see in traditional hard drives (or actually you don't see, since they hide it from you with the drive firmware).

      Anyway, in the short term, they would rather waste a GB or two and make the sizes the same as what you see in magnetic drives because the marketing is easier.

    3. Re:I'm an idiot by dextromulous · · Score: 1

      Not if the flash drive is comprised of 5 or 10 of these or something similar: http://www.linuxdevices.com/news/NS5845259932.html

      Sure, it may be convenient for chips/modules to be powers of 2, but not necessary in the case of flash.

      --
      There are two types of people in the world: those who divide people into two types and those who don't.
  12. Reason for using solid-state drives by pieterh · · Score: 4, Interesting

    Disclaimer: I paid the extra $1,000 for a SSD with my MacBook Air, so I'm probably biased, but most notebooks I've owned has had disk drive issues. It seems part of the price to pay for portable computing. Maybe I'm just brutal with them. The HDDs used in iPods seem more robust but they're slower than normal notebook drives.

    The main value of an SSD in a notebook is therefore that the notebook will last longer and there is much less chance of losing data due to disk failure.

    Additionally, SSDs are a bit faster, and they're silent and use less power. They are also a little lighter, I assume.

    On the down side, they're really expensive and writing files is slower so I guess you want to have lots of RAM and avoid swapping.

    In 3 years they'll cost 10% of what they cost today, and they'll be in more than 50% of notebooks.

    I don't see the advantage of SSDs in desktops, where it's trivial and normal to have full backups, and where power consumption, noise, weight, etc. are less important.

    So it's a little inaccurate to see SSDs as direct competitors to HDDs, ultimately they address two distinct markets, high capacity vs. high reliability. SSDs are always going to be for secondary computers, and portable devices. Of course it's also true that these compete with desktops.

    1. Re:Reason for using solid-state drives by DarkSkiesAhead · · Score: 1

      Additionally, SSDs are a bit faster, and they're silent and use less power. They are also a little lighter, I assume.

      On the down side, they're really expensive and writing files is slower so I guess you want to have lots of RAM and avoid swapping.

      Er, which is it?
    2. Re:Reason for using solid-state drives by GregPK · · Score: 1

      I'm looking forward to the day that we can just get rent movies from the store in a worm drive read only card format.

      Think, if you could get hd format movies in full length put onto an SSD that simply plays the movies. No more CD scratching.... Errors, etc.

    3. Re:Reason for using solid-state drives by vertinox · · Score: 2, Informative

      Technically SSD is both:

      Reads faster (ie boots quickly, apps open faster)
      Writes slower (ie files saves slower, page file churns sluggishly)

      --
      "I am the king of the Romans, and am superior to rules of grammar!"
      -Sigismund, Holy Roman Emperor (1368-1437)
    4. Re:Reason for using solid-state drives by nuzak · · Score: 1

      I've heard you can download movies off the tubes without having to go to a store. You can even do it without getting it from pirates, what with their parrots and eyepatches and all that. Amazons maybe.

      --
      Done with slashdot, done with nerds, getting a life.
    5. Re:Reason for using solid-state drives by maxume · · Score: 1

      Solid state is growing capacities slightly faster. If that keeps up, they will eventually be cheaper per unit capacity, and larger, making reliability and power consumption the deciding factors.

      --
      Nerd rage is the funniest rage.
    6. Re:Reason for using solid-state drives by ardin,mcallister · · Score: 1

      Both. Reading is faster. Writing, i cant remember the source, but i believe its a bit slower.

      --
      "Some men just want to watch the world burn..."
    7. Re:Reason for using solid-state drives by SuluSulu · · Score: 1

      I'm looking forward to the day that we can just get rent movies from the store in a worm drive read only card format.

      Think, if you could get hd format movies in full length put onto an SSD that simply plays the movies. No more CD scratching.... Errors, etc.
      Your expectations are to low. Forget getting out of my chair to go to the store! I'm waiting for full HD streaming over fiber.
    8. Re:Reason for using solid-state drives by ProfessionalCookie · · Score: 1

      I don't see the advantage of SSDs in desktops
      Speed. By the end of the year SSD will be faster, both reading and writing than HDD. There's some pretty hefty physical limitations on the spinning disk paradigm and everybody is really used to those. SSD doesn't have many of those limitations. Give it a year ;)
    9. Re:Reason for using solid-state drives by ProfessionalCookie · · Score: 1

      The Internet Wins.

      I would not be at all surprised if Blue-ray is the last distributed physical media for movie distribution.

    10. Re:Reason for using solid-state drives by chipace · · Score: 1

      "The main value of an SSD in a notebook is therefore that the notebook will last longer and there is much less chance of losing data due to disk failure."

      With cpu speeds and mobile gpu performance still increasing, why is it important to own a laptop that will last longer. 5 years from now a laptop could be much smaller, lighter and have much better performance. Your latest purchase confirms that.

      It's only when silicon performance and packaging technology slows down significantly, that longevity becomes important.

      Case-in-point: The Japanese government funded a rechargeable pacemaker development a number of years ago. Their initial design performed well, and the longevity was predicted at well over 10-15 years. The problem was that newer, non-rechargeable models had better features (distance telemetry, data storage, high speed telemetry, pressure sensors, accelerometers) that increased efficacy and saved money in doctor's visits & disease diagnostics. Their development was scrapped after that analysis.

    11. Re:Reason for using solid-state drives by DamonHD · · Score: 2, Interesting

      Hi,

      I'm here to quibble with "SSDs are always going to be for secondary computers, and portable devices."

      http://www.earth.org.uk/low-power-laptop.html

      I already use SSD (4GB SD card) as my primary Linux boot/main storage device to keep power consumption of my primary HTTP/SMTP/NTP/... Internet-facing server to under 20W. I also have a 160GB HDD, spun down as much as possible, for bulk data.

      If this 160GB drive had existed in the middle of last year when I speced the machine, I'd have had bought it like a shot to simplify life no end (and save a little more power). Laptop-mode - who needs it? (Actually it still might save a little power by batching and conflating operations, but much less I imagine.)

      Rgds

      Damon

      --
      http://m.earth.org.uk/
    12. Re:Reason for using solid-state drives by cleatsupkeep · · Score: 1

      Well it still could be HD-DVD... :-)

    13. Re:Reason for using solid-state drives by Nibs+Niven · · Score: 1

      "... in desktops, where it's trivial and normal to have full backups ..." If only that were even close to true...

    14. Re:Reason for using solid-state drives by Doppler00 · · Score: 1

      In the future there will be a HUGE advantage for SSD in low end desktops. There is a large fixed cost for all the mechanical parts of a hard drive. If you're developing a low cost consumer level PC for office/internet apps, having a single chip integrated on the motherboard could reduce costs significantly. The question is... how can the PC manufactures remove the bloat in their OS's to make this feasible?

    15. Re:Reason for using solid-state drives by GregPK · · Score: 1

      No one has really perfected an HD format for downloading yet.

      Quality is getting better all the time, storage requirements are going up with the addition of extra scenes, deleted scenes, director commentary, etc.

      I doubt Blue-ray is the long-term format going forward until they get it(layer it) up to around the Terabyte size with a read speed in excess of 100 megabytes a second for fast loading and menu navigation. I think the next step in video technology is going to be interactive like a mystery novel and/or a Holographic format where even more data is needed on a per disc basis which very much rules out Blue-ray as a going forward technology. It also rules out downloading as a video format technology, because the bandwidth constraints prevent any sort of instant on gratification.

  13. WOW!!! by sm62704 · · Score: 0, Redundant

    My PC at home has a combined total of 100 gb hard drive space, including HDa, C:, and D: and I'm nowhere near filling it up yet!

    I could have a computer without a hard drive and it would actually work!

    I wonder what these puppies are going to cost, anyway?

    --
    mcgrew's razor: Never attribute to stupidity that which can be explained by greedy self-interest
    1. Re:WOW!!! by Schnoogs · · Score: 0

      I have a 500 gig drive that is nearly full and starting next month I will be filling up a terrabyte with DVD data.

      My notebook has a 120 gig drive and between games, development data and multimedia it's almost full.

      In other words there is a market for a measly 160gigs! ;)

    2. Re:WOW!!! by crow · · Score: 1

      Here, too. Excluding media files, I don't see any use for more than 32GB on any of the systems that I use. But that's a big exclusion. My photos probably only take 10GB, but once you get into my music and my MythTV recordings, I'm pushing a TB, and I would have had to buy another drive if not for the writers' strike.

    3. Re:WOW!!! by sm62704 · · Score: 1

      Well sure, if all my MP3s were CD quality it would likely be full, let alone of I'd put my whole VHS and DVD library on it, even in compressed format.

      But you're right, there certainly is a market for computers with sub-200gb drives. Especially laptops.

      --
      mcgrew's razor: Never attribute to stupidity that which can be explained by greedy self-interest
  14. Re:I for one... by somersault · · Score: 1
    Governor Arnie called, he said:

    My CPU is a neural net processor, a learning computer.
    --
    which is totally what she said
  15. Re:I for one... by sm62704 · · Score: 1

    But what about a beowolf cluster of geometric rocks?

    --
    mcgrew's razor: Never attribute to stupidity that which can be explained by greedy self-interest
  16. Will never be cheaper by msgmonkey · · Score: 1

    The difficulty of fabricating flash memory are in the orders of magnitute more diffuclt compared to covering a metal disk with some magnetic material. So there will always be a market for magnetic media, unless that is replaced by some similarly cheap technology.

    1. Re:Will never be cheaper by jellomizer · · Score: 4, Funny

      Um sorry I was a sleep for 40 years and I just woke up... This Internet thing is pritty cool...

      But the Difficulty of fabricating Magnetic Memory is magnitutes more diffucly compared to punching holes in some cardstock. So there will always a need for Punchcards.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    2. Re:Will never be cheaper by msgmonkey · · Score: 1

      Erm, Magnetic Memory did n't get rid of Punchcards.

    3. Re:Will never be cheaper by jellomizer · · Score: 1

      Are you talking about in general or the fact that there is still useage of punch cards in some extent still existing....

      In gerneal computing Punchcards as they were used with early computers is dead. People do not store instructions on these cards for normal use (unless they are into retro stuff)

      Punch Cards are still used for storing small amounts of data, that can be handed to a person and then easilly put into a computer to track information such as Toll Roads without using EZ-Pass.

      I expect a simular fate for Magnetic Media, deligated to Bulk Storage and backups, But no one in their right mind will use it as the primary source of data. As well magnicic cards will still be in play.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    4. Re:Will never be cheaper by thanatos_x · · Score: 1

      I don't see regular HDDs going by the wayside in the near future. SSD does have several key advantages, namely reliability (no moving parts) and much faster random seek times. Sustained read times will catch up to HDDs eventually, or at least enough for the benefits of seek time to crush sustained read times.

      There will still be a reasonably large market for HDDs until the price/GB is within 125% of HDDs. Millisecond seek times are fine for HD movies, and they take up a huge amount of space. Eventually home media servers will become popular, and in that domain HDDs have a fairly large lead - SSDs weak points are those that HDDs excel at.

      As another comparison, tapes are still used to back data up - despite being very slow they're cheap.

      --
      I am not an expert. If I am misled in something, please correct me.
    5. Re:Will never be cheaper by slaingod · · Score: 1

      But burnable media will see a sharp decline in the near future when price per GB for HD's matches price per GB for burned media.

      --
      http://blog.slaingod.com
  17. Re:I for one... by MrShaggy · · Score: 1

    Does your rock run Linux?

    --
    I have mod points and I am not afraid to use them.
  18. yes but.... by sneakyimp · · Score: 1

    what SPEED are they? Anything higher than 150x?

  19. Re:I for one... by calebt3 · · Score: 1

    There is a difference between a solid-state overlord and an overlord with solid-state storage. My non-solid-state overlord does run Linux, though.

  20. Cool! ALL my p0rn on one flash drive by hoyeru · · Score: 0

    what will they think of next?

    --
    fuck karma, I like saying the truth better
  21. I asked this same question on LKML 6 months ago by rcb1974 · · Score: 5, Interesting

    Check our my post to the Linux Kernel Mailing List: http://lkml.org/lkml/2007/8/23/5 It drew a lot of responses from kernel developers.

  22. XYZ-byte Solid-State inevitable by davidwr · · Score: 5, Funny

    I'm here today to announce the future availability of 10TB solid-state drives.

    Pricing, manufacturing, and delivery date will be announced at a later date.

    --
    Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
    1. Re:XYZ-byte Solid-State inevitable by orielbean · · Score: 3, Funny

      It will ship with Duke Nukem.

    2. Re:XYZ-byte Solid-State inevitable by Shinmizu · · Score: 5, Funny

      He said terabyte, not yottabyte. You're jumping the gun a bit there.

    3. Re:XYZ-byte Solid-State inevitable by MrMacman2u · · Score: 1

      I scoff at your meager 10TB drives! Hah!

      I'm here to announce 100TB drives that can read and write at 10x the speed of your obsolete technology!

      Pricing, manufacturing and delivery date will be announced at a slightly later date than yours!

      --
      This signature is lame.
  23. Re:I for one... by Anonymous Coward · · Score: 0

    In my slightly buzzed state I just noticed your sig and initially started trying to translate it from leetspeek... *slaps forehead*

  24. year 2015 the end of the consumer hard disk? by backslashdot · · Score: 1

    It's quite possible that by 2015, most consumer PC's will not have hard drives. Hard drives will be relegated to servers that have over a terabyte.

  25. I'd rather have an SRAM drive instead of Flash ... by Skapare · · Score: 1

    ... where I can store my decryption key.

    --
    now we need to go OSS in diesel cars
  26. EduMacate yourself by geekoid · · Score: 1

    You can write continuously to a modern SSD for 12 years before wear is a factor.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  27. Indeed by SnoopJeDi · · Score: 4, Funny

    More proof that competing companies are good for consumers


    Quite so. I daresay this capitalism business is catching on rather quickly.
  28. I'd love to join in with the insightful comments by Rosy+At+Random · · Score: 2, Interesting
    But I'm going to have to settle for relaying the chorus swamping my mind:

    ... Want! Want! Want! Want! Want! Want! Want! Want! Want! Want! Want! Want! ...

    Damn, but I could do with a nice .ogg-compatible portable player with one of those in.

    OK, look, I'll try and say something worth reading: it has annoyed me quite a bit lately that, as SSD-driven audio players have mostly dominated over HDD ones in the last few years, the high-end of the capacity spectrum has become quite sparse; a few iPods that don't play .ogg and some very big and expensive media players that do. All I want is a nice, small, fairly inexpensive-ish ~100Gb .ogg player! Now, will someone please make me look like an idiot by telling me where to get one?

    --
    Would you like a slice of toast?
  29. Media? Also Scientific and Business data by Anubis350 · · Score: 1

    I'm in the middle of speccing 2 10TB fileservers for the research group I work for, and that's a pittance compared to some data storage needs in science and certainly in enterprise. Magnetic disc for large-scale storage isn't going anywhere for a while unless they can *really* push up SSD storage capacity cheaply.

    --
    "goodbye and hello, as always" ~Prince Corwin, from Zelazny's Amber series
  30. Re:I for one... by calebt3 · · Score: 1
  31. 100TB by Anonymous Coward · · Score: 0

    The only reason I didn't go with 100 is some smart-ass would suggestion a future technology would replace solid-state tech by then.

    Let me be that smart-ass.

    At 160GB, we are about 6 "doublings" away from 10GB, but a bit over 9 doublings away from 100GB.

    Assuming we double every 18 months or so, 6 doublings is about 9 years. In 9 years solid-state may very well be on its way out. By 2021 we should be using biological-state or something else rather than anything that resembles today's solid-state drives.

  32. I can't wait to see how this shakes up by QuantumRiff · · Score: 1

    Hopefully intel puts as much technical talent and care into the design of their Flash drives as they do into their totally kick ass intel graphics chips!

    --

    What are we going to do tonight Brain?
    1. Re:I can't wait to see how this shakes up by sobolwolf · · Score: 1
  33. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  34. Why only Laptops? by jez9999 · · Score: 1

    Whilst I can see the reason why laptops would be the primary beneficiary of this tech, where's my desktop flash-memory hard drive? I want one, damnit!

  35. Re:year 2015 the end of the consumer hard disk? by tecmec · · Score: 0, Informative

    I doubt that. Do you really think that in 7 years our internet bandwidth will be able to support that? Most US ISPs can't even handle a few bittorrent users, do you think they will be able to handle every used storing all their photos and videos somewhere else? ...that's not even taking into account things like the increasingly large size of computer games, etc. I have a hard time believing that we will move back towards a "server with many terminals" type type system any time soon. The cost of a personal computer doing everything locally is just too low.

  36. What about power consumption? by walterbyrd · · Score: 1

    Would a 160gb flash drive use more, or less, power than a 160gb SATA drive? Or, is it a matter of how much you use it? A drive like that would be awsome for portable devices, if the power consumption is much lower.

  37. Representational Difference? by lullabud · · Score: 2, Interesting
    I surely trust a man who thinks that moving a comma in an integer signifies a change in actual value. Surely.

    ...about 10,0000 write cycles in 1994, rising to 100,000 in 1997.
  38. SSDs Are More Reliable by gacl · · Score: 1

    Nowadays SSDs are more reliable than HDDs. I just checked the specifications for a Samsung SSD and a Toshiba HDD and they have, respectively, an MTBF of 2 000 000 hours and a MTTF of 300 000.

  39. Been waiting for this for years. by LinuxLuver · · Score: 1

    The first "large" RAM drive I ever saw was a 1GB shoebox that cost $300,000 that sat at the heart of one of Telecom New Zealand's number look-up services. A decade later we have 160GB flash drives that cost a tiny fraction of that. Good. I want one. :-)

    --
    Only boring people are ever bored.
  40. Nonsense. by samael · · Score: 1

    The HD in my Tivo was constantly writing for four years before I sold it (barring the odd power-down to move the box).
    The HD in my cable box is currently doing likewise.

    HDs can happily write constantly for extremely long periods of time.

  41. Re:year 2015 the end of the consumer hard disk? by torkus · · Score: 1

    ...my computer already has over a TB...

    --
    You can get rich if you own a politician, but you have to be rich to buy one in the first place.