Micron tests next-gen SSD

Micron tests next-gen SSD

Summary: Consumer NAND flash technology has serious problems in enterprise applications, leading the industry to look for better non-volatile semiconductor storage. Researchers at IBM have been characterizing just such a new drive from Micron.


While far from a new product announcement, a prototype SSD using a type of resistance RAM (ReRAM) - Phase Change Memory or PCM - as an alternative to NAND flash is welcome proof of industry progress. Because the flash used in today's SSDs has serious problems with reliability and scaling.

Flash in the pan
NAND flash used in SSDs today has a limited future for several reasons. As I noted two years ago in the The dismal science of flash:

NAND flash traps electrons in quantum wells. As feature sizes shrink, so do the quantum wells and, more critically, the number of trapped electrons that store the information. That makes it harder to preserve and read the stored information.

Thus the many contortions - wear-leveling, predictive ECC, DRAM buffers, over-provisioning - that SSDs employ to deliver performance and reasonable life. But the physics of NAND flash are brutal. Storage needs something better.

That something better looks to be ReRAM, which is currently implemented in a variety of technologies, including PCM. The invention of PCM predates flash, but it has been a long, slow slog. Indeed, despite multiple product announcements it isn't clear who, if anyone, is shipping product with embedded PCM.

Researchers at IBM Almaden used a Micron 64GiB PCM SSD based on 45nm technology for system level testing and simulation. They modeled a hypothetical storage device that used PCM SSDs, flash SSDs and hard drives, using an instrumented version of Linux to get nano-second accuracy of I/O latency.

This is important because chip-level performance stats may not reflect what is possible at a storage device level. For example, flash chips have very slow write cycles, but through the use of parallelism, smart garbage collection and DRAM buffers, that can be hidden in most apps.

PCM issues
PCM writes are non-trivial: a bit of material must be heated and allowed to cool for the material to reach either the amorphous or crystalline state that stores the data. This is a costly process, but once written the data is highly stable, unlike flash.

The modeled storage used PCM SSD for hot read-intensive data, flash SSD for hot write-intensive data, and disk drives for cold data. The simulations used storage traces from actual retail, bank and telco applications.

The researchers conclude:

Based on the results above, we observe that PCM can increase IOPS/$ value by 12% (bank) to 66% (telecommunication company) even assuming that PCM is 4× more expensive than flash.

The Storage Bits take
I hope to learn more at next week's File and Storage Technology (FAST '14) conference in Silicon Valley, where the not-yet-publicly-available paper Evaluating Phase Change Memory for Enterprise Storage Systems: A Study of Caching and Tiering Approaches by Hyojun Kim, Sangeetha Seshadri, Clement L. Dickey and Lawrence Chiu of IBM Almaden Research will be formally presented.

But on the face of it this is not a persuasive case for PCM. The performance improvements described and the economic benefits derived fall short of those required, in my experience, to drive commercial adoption.

But the real point is that the industry is actively looking beyond today's flash to tomorrow's underlying storage technology - a technology that, like PCM, should be much more robust than NAND flash. Flash has the considerable economic advantage that comes with enormous volume, but that came only after decades of research and development.

While PCM may not win the race for the next-gen flash, this research gives a more-nuanced approach to how a more robust - and more costly - ReRAM could be incorporated into significant commercial products. It is this kind of decades-long development that creates the next "overnight" sensation in technology as well as life.

Comments welcome, of course. An example of the creative incorporation of a new storage technology can be seen in the architecture of Nimble Storage, described in a video white paper I did for them almost three years ago.

Topics: Storage, Emerging Tech, Hardware

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Where are the memristors HP?

    Not rocket science.
    Johnny Vegas
    • I think

      they're renaming "memristor" to "phase change memory," since that is the physical mechanism by which the memristor action occurs.
      Jacob VanWagoner
      • Jacob, I wanted one too

        Just installed a 120g that I purchased for under $100. Several available in market. Most my data is on 1TB standard drive, system alone on the SDC.

        My startups/shutdown are superfast. Programs start in a flash (pardon the pun). I think its the best of both worlds.

        99% of my data is "cold" .... Best improvement I could have made to my PC.
      • Only very slight similarities in principle.

        Check Wikipaedia, for example.
        They are both resistive technologies, but memristors operate bi-directionally.
        In the HP type of memristor there is an hypothesized partial phase change that confers energy isolation to the two states, and, as with phase change, is probably the major technical hurdle, since it requires high current density and high temperature, though with memristor, the current density and temperature are not as extreme.
    • Re: Where are the memristors HP?

      They've been out for a few years, as promised. Haven't you noticed?
  • I'm hoping

    that the introduction of a technology touted as better than NAND Flash will push the price of Flash memory down. I have wanted SSDs for my computers for a few years now, but they've been too expensive so far.
    Jacob VanWagoner
    • No hope that the new technology will be cheaper

      Jacob, the flash guys are doing a good job bringing the costs down, IMHO. If you shop sales you can get a quality flash drive for 50¢ or less per gigabyte. My first USB flash drive was $1 per megabyte 13 years ago.

      Any new technology will be more costly than flash because it won't have the volume that flash has today. That's why the IBM guys were looking at PCM as a leverage technology rather than a replacement technology.

      R Harris
      • I agree, price drops are key

        Remember when a 40-inch plasma HD-TV was $40,000? Few were sold. When a new technology provides a compelling reason to get it at, AND it's available at a "reasonable" price, that's when consumers stand up an notice.
    • The most tantalizing thing about memristors is the potential for AI

      See this recent article:

      Also, an early demonstration HP did on AI with memristors:
      • Re: potential for AI

        Of course, don't let the government know about this!
  • Isn't PCM used for rewritable CDs & DVDs?

    I understand that PCM is currently in use for rewritable CDs & DVDs. Or am I mistaken? What I'd like to know is how any PCM device is going to heat and cool the individual bits and how small can the individual bits be? With a rewritable optical disc, the laser does the heating and cooling is aided aby the air sliding past the spinning disc. I remember working alongside colleagues in the 1970s in Plessey who developed holographic storage. The only problem then was that there was no demand for the enormous capacity that holographic storage offered. Maybe that's another option.
    • Re: Isn't PCM used for rewritable CDs & DVDs?

      Yes. Computer memory type PCM changes state using heat generated by electrical resistance.
      The disc technology uses lasers to generate the heat.
  • That PCM is a phase-change chemical dye

    Totally different tech.

    R Harris
  • I'm confused.

    The principal problem with NAND flash memory is not stability. It's the gradual loss of the ability to re-write the memory.
    • The two are related

      The more writes to a cell the shorter the time the cell can retain the written data. If you only write once the data retention is very good. But as cell wears the retention time drops. When you can't write anymore you still have the last written data, but not for long.

      All in all, a pretty rickety medium.

      R Harris
  • I'm still betting on Bubble Memory!

    Only you "old school" folks will get this reference
    • LOL

      I haven't heard the words 'Bubble Memory" since the early/mid 80's and my friends and I had Apple II E's.
  • Good bet.