Slashdot Mirror

← Back to Stories (view on slashdot.org)

Intel, Micron Boost Flash Memory Speed by Five Times

Posted by ScuttleMonkey on from the fastest-draw-in-the-valley dept.

Lucas123 writes "IM Flash Technologies, a joint venture between Intel and Micron, announced they've been able to improve NAND memory and its circuitry in order to boost read/write speeds by five times their current ability. The new 8Gbit single-level cell, high-speed NAND chip will offer 200MB/sec read speeds and write speeds of up to 100MB/sec, which means faster data transfer between devices like solid-state drives and video cards. IM Flash Technologies plans to begin shipping the new chip later this year."

22 of 67 comments (clear)

  1. That's fast by Drinking+Bleach · · Score: 2, Interesting

    Will it finally make sense for USB 3 flash drives? ;)

  2. AVAILABILITY? by Anonymous Coward · · Score: 2, Interesting

    So what are the read/write cycles, how much will they cost, and when can I get 200GB of them all in a nice pretty box? Even 10GB would be good for a nice little web server. Near zero latency would mean slashdotting is reduced to network bandwidth.

  3. Re:Video cards? by repvik · · Score: 4, Informative

    Yes, video cards need fast RAM. If you haven't noticed, this article is about flash memory, not RAM. If you shove this crap into a video card, you'll be going a helluvalot slower than you are today ;)

  4. Re:Faster USB needed by timeOday · · Score: 4, Insightful

    Yeah, but for solid state hard drives this is quite a leap. I'm starting to think winchester drives are going to be extinct within 5 years.

  5. Re:Faster USB needed by ciroknight · · Score: 3, Informative

    You're right. If only there was a new, faster USB standard that would be able to take advantage of these new data rates. They could call it "USB 3.0", or "USB SuperSpeed" or something. Oh Wait...

    --
    "Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
  6. Re:Video cards? by blahplusplus · · Score: 3, Insightful

    Even with the bandwidth increases, what you're spouting is nonsense. The local bandwidth of ram is infinitely faster and cheaper then flash, period. Not to mention: It doesn't wear out. Why would you put flash in a video card which does insane amounts of reads and writes per second? You'd have to be an idiot.

  7. Re:Video cards? by Erpo · · Score: 4, Funny

    The blurb says this will lead to faster transfers between flash memory and video card memory, which is certainly true. Faster flash will lead to faster transfer between flash and just about anything.

    I think this is similar to how the latest greatest processors are marketed as improving the "internet experience." Well sure. Not having a CPU at all or having a CPU from twelve years ago will hamper your "internet experience" compared to any new CPU.

  8. Re:Faster USB needed by Dr.+Spork · · Score: 3, Insightful

    They won't be extinct, but they'll be used for storage. If I'm booting my operating system from a spinning disk in 2013, I'll be pretty disappointed with technology!

  9. What about lifetime? by Z00L00K · · Score: 2, Interesting
    The lifetime of a flash memory has been one of the issues with flash memories. How good is that on those memories? Will they die after 1000 cycles or after a billion cycles?

    As usual - the lifetime of a product also requires the consumers to buy a new hot version.

    --
    If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
    1. Re:What about lifetime? by eebra82 · · Score: 2, Insightful

      Well, consider this: NAND is commonly used in solid state drives. I doubt companies like Dell, Lenovo and Apple would sell computers configured with SSD:s if they sucked it down with only a few cycles. This was a problem in early versions, but things have improved much and will surely improve to a point that makes it practically "unbreakable".

      --
      Full Tilt
    2. Re:What about lifetime? by tlhIngan · · Score: 2, Interesting

      Well, consider this: NAND is commonly used in solid state drives. I doubt companies like Dell, Lenovo and Apple would sell computers configured with SSD:s if they sucked it down with only a few cycles. This was a problem in early versions, but things have improved much and will surely improve to a point that makes it practically "unbreakable".


      Not really.

      There's still cycle time limits. The main issue came from NOR flash, which is different from NAND. NOR flash came first (mid-80s), and the very early versions suffered from poor lifetimes (~10,000 write-erase cycles). (However, they're perfect for firmware, which is their initial purpose - even during development, it's rare to wear it out). Modern NOR flash has a cycle life of around 100,000 cycles.

      In the early 90s, NAND flash came out, and due to their exploitation of quantum mechanics (NOR flash uses tunnelling and hot-electon injection (literally forcing electrons through the insulator). NAND flash uses tunnelling exclusively) resulted in a significant improvement in life - normally 1,000,000 cycles.

      Add in wear levelling, and things get interesting. Assuming a perfect wear-levelling algorithm, and maybe a large-block NAND chip (128kB block size), a 128MB chip (tiny these days) has 1,024 blocks. To wear it out, requires over a billion write-erase cycles! A GB chip would have 8192 blocks, thus over 8 billion write-erase cycles. And you want 32GB/64GB SSDs? It's gotten to the point where an SSD in normal use will probably outlast a mechanical disk.

      Oh, and most flash chips, these cycle times are very conservative - most will survive another order of magnitude of erase-write cycles before becoming unusuable.
  10. Some interesting possibilities open up. by jd · · Score: 3, Insightful
    LinuxBIOS was already down to 3 seconds, and I'll guess that slow flash access times contributed some to that. We could also see the revival of cartridge hard disks, only solid-state. A variant on that would be to have a RAID array where one or more of the disks were replaced with flash devices. In either case, you'd probably improve longevity and definitely improve resilience to things like shock. It wouldn't be cheap, sure, but the people mainly concerned with ruggedized technology (aircraft vendors, the military, war correspondents) are less likely to be concerned by price than by whether it'll survive the environment.

    It could significantly increase the usefulness of suspend/resume at the OS level. The limits on writes is a headache, but it would be possible to treat flash devices as additional swap space, making it theoretically possible to have hot-swappable swap devices as per some rather ancient mainframes. (Virtual swap space can be larger than the physical space directly available to a machine.)

    --
    It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
    1. Re:Some interesting possibilities open up. by Bloater · · Score: 2, Informative

      LinuxBIOS was already down to 3 seconds LinuxBIOS is now called coreboot
    2. Re:Some interesting possibilities open up. by Tumbleweed · · Score: 2, Funny

      We could also see the revival of cartridge hard disks, only solid-state.

      You watch - soon we'll all be storing data on little coloured cartridges ala Star Trek.

      Ladies, get your beehive hairdos and miniskirts ready, Kirk is in the hizzy!

  11. Re:Wonder when... by Bender_ · · Score: 2, Interesting


    AMD divested of their memory business years ago. You should look more often.

  12. Re:Faster USB needed by Kjella · · Score: 2, Interesting

    Yeah, but for solid state hard drives this is quite a leap. I'm starting to think winchester drives are going to be extinct within 5 years. Not unless the price still comes waaaaaaaaaaaaay down. It's not a technical difficulty of making it, I think BitMicro or whoever it was showed off a 900GB SSD in a 2.5" form factor. I don't remember the read/write specs but I think those too beat the crap out of any normal HDD. That trumphs everything a regular disk does, even on capacity. The downside is price, looking at the SSDs available now there's a 100:1 premium per GB. That might work out in a business laptop where 32GB SSD is more useful than 320GB HDD (which is about 3x as expensive as bulk desktop storage anyway), but not for normal users. Remember as laptops take over, more and more people only have a laptop and so they want to store everything on it. Heck, just game installs are 5GB each these days. If you use a digicam regularly, that adds up to some gigs particularly if you store in raw or edit in photoshop and save losslessly. That mp3 collection for your iPod, also a lot. But yeah, I want a quiet disk to boot from and my file server to be far, far away. Unfortunately there's no far, far away in this apartment...
    --
    Live today, because you never know what tomorrow brings
  13. filesystem by yupa · · Score: 2, Insightful

    This is great, but we realy miss filesystem for such big NAND.
    Either we use FTL [1](flash translation layer) to put FAT, but that that's quite ugly (FAT is not aware of flash and not robust to power lost, FTL is optimized for FAT).
    Either we put flash filesystem like jffs2 or yaffs2, but they will eat lot's of RAM and take lot's of time on such big flash.

    I wonder what are the performance with a filesystem.

    PS : there is logfs or ubifs that should be better flash fs, but there are not ready.

    [1] BTW FTL is patented.

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

      Most flash devices now spread writes to the device using the controller. There is no real need anymore for those kind of file systems. Yes, current devices use FAT. Anything supports FAT and it is easy to implement in small devices (e.g. camera's, phones etc). But you can just as easily put in NTFS, or EXT2/3.

  14. Re:Wonder when... by Circle+of+Owls · · Score: 3, Informative

    AMD had a flash memory division; it's flash ran alongside their logic products in Fab25, but took a backseat to them in both production and engineering. As a result, their flash technology rapidly slipped behind the market leaders. AMD then formed a joint venture with Fujitsu called Fujitsu AMD Semiconductor Limited (FASL) to jointly develop and market their products at about the same time that AMD was moving their logic line to Fab30. FASL was soon split from both companies into a separate entity, and renamed to Spansion. Spansion has been making significant progress to regain both the market share and technology that AMD's priorities lost for them.

  15. Re:When it rains, it pours by ColdWetDog · · Score: 2, Informative

    One place this may really help is cameras. The shutter lag is still bad, and this might help.

    If it's shutter lag that bothers you, get a decent camera. Today. Flash RAM isn't the problem here. ANY DSLR made in the past five years has quite acceptable shutter lag for most people. The higher end models have shutter latencies better than any "normal" camera ever made. There are even a couple of point & shoots with reasonable speeds. Check out the reviews on DP Reviews.

    Happy snapping.

    --
    Faster! Faster! Faster would be better!
  16. Re:Video cards? by Yvan256 · · Score: 2, Funny

    1. Make videocard which uses Flash
    2. Advertize the hell out of it
    3. People buy videocards that last about a month at most
    4. Profit!

  17. Re:Give me better sekk times by tlhIngan · · Score: 2, Informative

    Try do random access on the next USB-flash stick you have access too.


    That's a USB issue, not a flash issue.

    The reason is that USB does things in transactions, and has to schedule all the transactions with priority. This is because of USB's fundamental flaw - it requires the host to poll devices. So a host will poll interrupt devices first, then handle isochronous transfers (bandwidth and time dependent traffic). Leftover bandwidth is then allocated to control and then bulk traffic. A USB host can do this once every millisecond, but most OSes break it out into more coarse granularity to avoid overloading the CPU when doing USB transfers. 10ms is about average for Windows, Linux is around 4ms. Basically, Windows will schedule all traffic on 10ms boundaries, so every part of a transaction will take place every 10ms. (10ms is a nice number because it means Windows can do the scheduling every timer tick).

    If you do a USB disk request (read block N), the USB Mass storage driver will make a transaction to read a block. It will then issue the request to Windows' USB stack, which then add it with all the requests. If there's sufficient bandwidth in the next 10ms frame, it'll add the request to that frame. In the meantime, it's handling the current frame. When the next frame goes through, it sends the request, and if your USB stick is fast enough (usually is, but hard disks, it isn't) it responds immediately. If your USB stick isn't fast enough, then it will accept the request and wait for Windows to poll it again to see if there's any data, at which point the data will be transferred.

    It's not the flash memory doing the seek (in fact, every time you access it, you "seek" it, it's part of the normal behavior) it's USB.