Slashdot Mirror

← Back to Stories (view on slashdot.org)

Researchers Invent a Way to Speed Intel's 3D XPoint Computer Memory (ieee.org)

Posted by msmash on from the new-innovations dept.

Memory modules using Intel's 3D XPoint technology are on their way, and researchers in North Carolina have already figured out how to make them better. New submitter mnemotronic writes: At the 45th ICSA (International Symposium on Computer Architecture), a group of researchers from North Carolina State University led by Prof. Yan Solihin proposed a method called lazy consistency to speedup write operations to 3d XPoint memory. XPoint, developed by Intel and Micron, is non-volatile, cheaper and denser than DRAM but requires more power and writing takes longer. The method proposed reduces write overhead times from 9% to 1% by incorporating a checksum to the cache memory system. The researchers were not able to verify their approach on actual XPoint memory, as those products only recently started sampling. They tested using simulations and DRAM and plan to verify when Intel's modules become more widely available.

7 of 43 comments (clear)

  1. Re:MRAM vs...? by Forever+Wondering · · Score: 2

    Probably. IIRC ...

    MRAM consumes less power than DRAM (vs. more). MRAM is _faster_ than DRAM (and is as fast as L2 cache).

    It also has a very small bit cell size (so very high density).

    So, it beats out 3D-XPoint (aka Optane) on almost every point.

    Also, MRAM doesn't "wear out". From what I've read, 3D-XPoint is better than flash on this, but, eventually, has a wear out point.

    --
    Like a good neighbor, fsck is there ...
  2. Won't Work by sexconker · · Score: 4, Interesting

    The delay is because XPoint doesn't work. The writes usually take, but sometimes they don't. Intel hasn't figured out why.

    They current practice is to verify all the writes and simply redo them if they don't take.
    This means you're tying up the the bus, and this is why Intel now recommends dedicating entire memory channels to XPoint instead of mixing and matching with DRAM. If you have XPoint in all of your channels, your latencies go through the roof and your performance tanks.

    Wait for generation 3 before considering XPoint NVDIMMs.

    1. Re:Won't Work by swb · · Score: 2

      Why would you mix and match DRAM and Xpoint on the same bus anyway when Xpoint is so much slower than DRAM? Even without extra verification and writes its still much slower than DRAM and would seem to clog the channel.

      I'll admit that maybe I don't know something about existing DRAM access paths/channels/buses, AFAIK the NUMA node was basically the smallest subdivision. Or is it possible to address individual DRAM modules/pages in parallel with others on the same NUMA node?

      I guess I had figured that DRAM interfacing Xpoint was meant to be mostly a speed/simplicity thing, faster than PCI NVMe and meant to be on a distinct memory bus from DRAM.

      The application that came to mind was just using it as superfast block storage in maybe hyperconverged systems or monolithic DB servers.

    2. Re:Won't Work by DigiShaman · · Score: 2

      Would it not make sense to just create an NVMe SSD with Xpoint as the internal cache of the drive for the rest of the NAND memory in a single storage device?

      --
      Life is not for the lazy.
    3. Re:Won't Work by Junta · · Score: 2

      Note you can make numa domains smaller. It is quite plausible that 3d xpoint in 'memory' mode appears as a different NUMA domain with SLIT indicating higher penalty for use. NUMA is relatively expressive and can describe the high level divisions and implications of this sort of approach.

      A challenge for Intel is justifying sitting in a rather inconvenient DIMM slot. It may be able to deliver better performance rather than PCIe,, but even PCIe attached NVME has had a very protracted adoption cycle as the relative value of going from SAS/SATA SSD to PCIe attached SSD is not as dramatic (it's also debatable whether they needed PCIe so much as they needed NVM, SATA/SAS protocols baked in assumptions that do not apply to a highly parallel storage architecture, NVMe provides an alternative protocol with, for example, thousands more queues each thousands times deeper than anything in SATA or SAS).

      --
      XML is like violence. If it doesn't solve the problem, use more.
  3. Re:lazy consistency by radarskiy · · Score: 2

    Speculation is about resolving predicates. Consistency is about resolving dependencies.

  4. Re:No cited article by Solandri · · Score: 3, Informative

    Ever since the Slashdot redesign a few years back, the main link to TFA appears to the right of the title, where it's easy to miss and not at all obvious that it's a link. But I imagine they paid some designer handsomely to make the site less usable.