Forgot your password?
typodupeerror
IBM Hardware

IBM Creates Multi-Bit Phase Change Memory 82

Posted by samzenpus
from the look-what-we-found dept.
Lucas123 writes "In what is likely to be a strong rival to NAND flash memory, IBM today announced it has been able to successfully store more than one bit of data per cell in a more stable non-volatile memory called phase-change memory (PCM). Unlike NAND, Previously, PCM couldn't contend with flash because of its low capacity points. PCM does not require that data be erased before new data is written to it, which reduces write amplification or wear out and it has 100 times the write performance of flash. IBM researchers say they plan to license the technology to memory manufacturers instead of producing it themselves."
This discussion has been archived. No new comments can be posted.

IBM Creates Multi-Bit Phase Change Memory

Comments Filter:
  • Dad, tell us again about how you used to store your data on spinning disks....
    • Zero moving parts, here we come.

      Let me know when we can get fans that don't need blades. No, not that hoop thing. It has blades in the base.

    • by ByOhTek (1181381)

      That's nice. The only nail that counts, will be the one where solid state is at least modestly cheaper for a given amount of space.

      Mind you, I look forward to that nail, but until it gets here, it's not yet time to party.

      • I don't think it needs to be cheaper at all. Just cheap enough. I like having no moving parts in my netbook, tablet, phone etc. I also don't have any need for more than say 300GB to be comfortable on anything but a media PC right now.

      • Even if it stays more expensive, if solid state can get close (within a factor of 2, say) to the same cost for the same storage space it'll take off like wildfire given the tremendous speed and reliability advantages.

      • That's nice. The only nail that counts, will be the one where solid state is at least modestly cheaper for a given amount of space.

        Mind you, I look forward to that nail, but until it gets here, it's not yet time to party.

        For most users, disk storage is already much larger than needed. All that is needed is that an SSD have cheaper unit cost while still providing adequate storage. Even if it is still 4x the cost per byte, it won't matter because the extra storage isn't useful.

        • For most users, disk storage is already much larger than needed. All that is needed is that an SSD have cheaper unit cost while still providing adequate storage. Even if it is still 4x the cost per byte, it won't matter because the extra storage isn't useful.

          640K is more than enough for the average user too....

          The problem is you never know how the average user will use that disk space. I found a computer in the trash a few months ago, and it was fill with crappy music and pictures some some kids. Over 300Gigs worth of the stuff, with I formatted to install something useful like Linux.

          • by yarnosh (2055818)
            I don't think he was predicting the future with that statement. I think he was specifically talking about now. Obviously this solid state technology would get cheaper and storage would become greater over time. It would just be a bit behind spinning disks initially. For me SSD is great for my laptop. Installing it was a huge performance boost and I don't need the same amount of storage as my desktop/media/gaming PC where spinning disks work fine. Though ultimately it would be nice to quiet that sucker down.
    • by jewelie (752077)

      Ironic given IBMs role in their invention!

    • Maybe after it comes out. In the mean time, I imagine the cloud, tablets, and smartphones are doing more to kill the disc.

      • by wagnerrp (1305589)
        Someone is going to have to have the disks somewhere. The individual user may only have a single SSD in their device, but someone needs to provide the bulk storage.
        • So does 'nail in the coffin' mean "nobody ever uses discs ever again period end of story they're completely extinct" or does it mean "not really a household item anymore, like vinyl"?

          I'm guessing the answer is: "The opposite of what you intended so I can appear smart by correcting you."

          • by TheLink (130905)
            Hmm are you tempting him to do it again with "vinyl"? ;)
          • by wagnerrp (1305589)

            If I wanted to appear smart, I would correct you a second time on the use of disk, since you missed it the first time. Disc is from the Greek discus, a later adopted spelling of the term, and used to describe things you throw, like frisbees and CDs. Disk is the more traditional English spelling, and used to describe storage devices, since that is the spelling IBM chose to use when they started manufacturing disk drives 55 years ago.

            I would consider this a nail in the coffin if it resulted in a drastic dec

            • If I wanted to appear smart, I would correct you a second time on the use of disk...used to describe things you throw, like frisbees and CDs.

              CD's?

              Another word for disc is 'platter'. And since we're talking about solid state taking over....

              ...due to its basic design, they're rapidly closing in on a wall, past which they cannot miniaturize it any further...

              Yeah, unlike magnetic drives...

              If the average consumer stops storing stuff locally, and instead migrates to the cloud, hosting companies are going to have all this data they need to store, with only modest performance requirements. The most economical way to achieve this for the foreseeable future will be disk drives.

              That's an important detail, iddn't it? Another important detail is that having lots of people store their same information on the same network means serious efficiency gains can be made. And yet another important detail is that laptops are increasingly becoming primary machines, but they typically only have 1/4th the storage available to them their desktop counterparts have. W

      • by glwtta (532858)
        And how do you imagine "the cloud" (and by extension phones and tablets) stores its data?
        • You'll have to explain to me how consumers moving their data to the cloud will mean an increase of hard disc sales.

          • by glwtta (532858)
            You'll have to explain to me how consumers moving their data to the cloud will mean an increase of hard disc sales.

            Well, it's pretty involved: consumers will move their data to the cloud, at which point cloud providers will need disks to store that data.

            Storage utilization is likely to be more efficient, so I don't know if that will actually increase sales, but then again, we were talking about "killing the disk" not "increasing sales". The cloud, being essentially a large collection of disks, seems
            • ...but then again, we were talking about "killing the disk" not "increasing sales"...

              No... we're not. I said 'do more to kill the disc'.

              • by glwtta (532858)
                No... we're not. I said 'do more to kill the disc'.

                Ok, you lost me; when you said do more to kill the disk you were not talking about killing the disk? Or are you implying something clever with your alternative spelling?
                • Ok, you lost me; when you said do more to kill the disk you were not talking about killing the disk? Or are you implying something clever with your alternative spelling?

                  Okay... do you understand the difference between slowing down and stopping?

                  • by glwtta (532858)
                    Wait, wait, I know this one!

                    "Stopping" is the one you said, and "slowing down" is the one I was supposed to intuit you meant, right?

                    Which isn't true either, but, whatever.
                    • "Stopping" is the one you said...

                      No.

                      *sigh*

                      "IBM Creates Multi-Bit Phase Change Memory...

                      Another nail in the Coffin of the Hard Drive. Dad, tell us again about how you used to store your data on spinning disks....

                      Maybe after it comes out. In the mean time, I imagine the cloud, tablets, and smartphones are doing more to kill the disc."

                      So you're saying that the conversation above reads as: "the cloud etc. will bring the hard disc to a sudden and dramatic end", right?

          • by Changa_MC (827317)

            You'll have to explain to me how consumers moving their data to the cloud will mean an increase of hard disc sales.

            No, you'll need to explain how consumers moving their data to The Cloud will mean an decrease of hard disc sales, since that was your previous claim. The Cloud is remote servers storing data on spinning disks. So consumers storing data at home as well as data in The Cloud means more hard drives in use, not less.

            • No, you'll need to explain how consumers moving their data to The Cloud ... ...So consumers storing data at home as well as data in The Cloud means more hard drives in use, not less.

              See how you used the word 'move' there in the beginning? You already understand my point.

        • Magic pixie bits. The challenge is capturing enough pixies to feed the industrial grinders.
        • by Yvan256 (722131)

          In another cloud?

    • SSDs remain what, 1/6th the capacity and about 25x the cost per GB of spinning platters? Yea, thats a lot of nails in the HDD coffin alright.

      Last I checked, a 2TB drive could be had for $80, which is $0.04 per GB, and you can scale to 3TB if you want. A 256GB SSD drive retails for what, $200? $300? Thats $1.00 per GB. I dont know what magical world you live in where that is suitable for every use case, but its certainly no good for large data storage, particularly when there are questions regarding rel

      • by wisty (1335733)

        Is that server grad? Because servers usually use SAS, which is about $300 for 300GB. Or roughly the same $/GB as SSD.

        Or did SSD kill them?

      • particularly when there are questions regarding reliability and what real-world flash failure looks like (is it still readable? Does the controller start spewing garbage?)

        CodingHorror answers this [codinghorror.com]:

        1. They fail. A lot. Within months. If you're lucky.
        2. The failure is they just die completely and conclusively.

        However, they stress the important point:

        3. They're so ridiculously fast that you want one anyway. And you'll keep replacing it when, not if, it fails.

        For server use, the current fashion is as a gigantic cache in front of magnetic hard disks.

        For either desktop or server, "GET REALLY GOOD AT BACKUPS, YOU'LL USE THEM" is the thing to remember.

    • by dbc (135354) on Thursday June 30, 2011 @01:49PM (#36624970)

      "Dad, tell us again about how you used to store your data on spinning disks...."

      Who here has stored a program on punched paper tape using an ASR 33 teletype? *raises hand*

      • by ahadsell (248479) on Thursday June 30, 2011 @02:07PM (#36625260)

        Also *raises hand*.

        On one system we stored programs by wiring them into a ROM. By hand. One wire per word, wrapped around the center pole of the E-cores clockwise for a 1, or counterclockwise for a 0. Then solder one end of the wire to the correct X address, and the other end to the correct Y address. Total, 256 16-bit words per board (Z was decoded to board-select).

        Yes, I am old.

      • Also raises hand. On one project I loaded the software, powered down, pulled the memory card with its core memory, walked though a food processing plant, plugged it back in, and started debugging. It was quicker. The customer insisted on receiving a paper tape copy and only had the printer on the tty. 40+k used almost an entire roll. I don't think anyone ever tried to read it back in.
      • by bityz (2011656)
        raises hand... but some of you guys are making me feel young :) (wrapping your own cores?)... somewhat after my punched tape/teletype experience, I built a PC (back then, "built a PC" meant soldering) and eventually moved up to a "stringy floppy"... I think the storage progression in my life went: paper tape, 5 1/4" floppy, stringy floppy, massive 4'x2'x8" 500M harddrive array, 2G SD card. I may have skipped a few steps there :)
  • I this story went up as I hit submit on it.
  • by goruka (1721094) on Thursday June 30, 2011 @01:22PM (#36624600)
    we all know researchers don't take business decisions
    • by mikeru22 (1222780)

      we all know researchers don't take business decisions

      haha. Not the real researchers, anyway.

    • by gl4ss (559668)
      *they plan to license the technology to memory manufacturers instead of producing it themselves* well, they don't know how to actually do it en masse in the way they're proposing. if they could, they sure would do it and be selling it at 100x the price of normal memory for enterprise use, today.. but they don't have yet what they're promising the tech could do - by manufacturing companies...
      • I don't think IBM is declining to actually manufacture the chips because they don't know how; rather it's because IBM doesn't have fabs large enough to make the chips in sufficient quantities. IBM's fabs are sized for CPU production, not mass-scale memory chip manufacture.

  • by Anonymous Coward

    they plan to license the technology to memory manufacturers instead of producing it themselves

    This is code for 'send it to Asia.' HP is doing the same thing with their memristor based ReRAM, licensing it to Hynix. Guess we're all done building chip foundries anywhere in the West. As goes your manufacturing base, so goes your technology, just like we we're told [slashdot.org] would happen.

  • I like flash, in fact I just bought some SSDs, but it does have some problems that really need addressing long term. This stuff looks like maybe it'll be the solution. Also it looks pretty workable, it isn't pie in the sky. According to the article, PCM already exists and is in use. This is just an improvement on it. They also claim widespread use by 2016 of the new tech.

    So here's hoping, this looks like it may be what we need to really kick off the move to solid state storage.

    • by poetmatt (793785)

      I have a feeling we might end up with flash+PCM combinations in some form.

    • by Caratted (806506) *
      Meh. New tech is all well and good, the problem with SSD is the cost of manufacturing. The same thing occurred with high capacity spindle drives in the late-90's. The tech to produce them cost a lot, so the platters cost a lot. When the supply goes up, the price goes down.

      I don't have anything against IBM, but the only thing contributing to the high-cost of SSD and its relatively low usage is the fact that there aren't billions of drives on the market. When there are, the machines to manufacture them
      • by Chirs (87576)

        the only thing contributing to the high-cost of SSD and its relatively low usage is the fact that there aren't billions of drives on the market

        Sorry, but fabbing flash memory is a lot more resource-intensive than making disk platters.

    • by yarnosh (2055818)

      I like flash, in fact I just bought some SSDs, but it does have some problems that really need addressing long term.

      Maybe in theory, but my OCZ Vertex is awesome. I know the cheap SSDs can be a bit flaky if pushed to limits, but Intel and OCZ make some fine SS drives.

  • ...much more efficient swap space! :) This sounds great.

    I know it doesn't need to be erased in bulk, but does it need to be written in bulk, or does it have limited read or write cycles?

    • Truely the "Holy Grail" of storage is to go back to core memory. No, not the kind that used ferrite rings. Rather, a very fast type of unified L cache that represents both RAM and non-volitile storage. It provides the kind of hardware platform DB admins dream of.

    • by yarnosh (2055818)
      If you're hitting swap on a regular basis, you need more RAM. Swap is just a backup on most systems.
  • It's density+cost per bit that will change things. PCM is still orders of magnitude slower than SRAM and DRAM and the Memristor (HP) may still beat it to market for the aforementioned parameters. Nice that so many irons are in the fire to retire spinning rust, can't happen soon enough for me.
  • Am I the only one that saw the "up to 5 million write cycles...enterprise usage of 3,00 write cycles per hour" as a pretty fail example? By their own statement that's 70 days of life in that use case....
    • by yarnosh (2055818)
      I'm skeptical of that 3,000 figure. Is that an average over all the sectors on the drive? Or is just the hottest of sectors on the drive? Seems unlikely that you'd rewrite the full drive 3,000 times every hour. That's insane, if not impossible. I'm going to assume that's a single sector being reweritten 3,000 times in a hour. With flash you use write leveling to spread that all the sectors. I don't think it is as big a deal as it sounds. 5 million writes per sector (or cell) is a LOT.
  • by Anonymous Coward

    IBM researchers say they plan to license the technology to memory manufacturers instead of producing it themselves.

    Yet another obscure company who sells licenses but doesn't make products, so they never have to pay anyone else or cross-license. Whoever heard of this "IB--" oh. Oh, them.

"A great many people think they are thinking when they are merely rearranging their prejudices." -- William James

Working...