SAN JOSE, Calif. – HP Labs continues to make progress on its long term vision of smart memories based on its memristors as alternatives to server CPUs. The devices are one of several new categories of chips likely to seize the moment of disruptive changes in data center technologies and workloads ahead, said an HP Labs researcher.
Separately, Hewlett-Packard Co. is expected to announce within weeks the next steps in its Project Moonshot, its work on ARM- and Atom-based servers. HP is working with a broad group of companies including processor providers AMD, Applied Micro, Calxeda, Cavium and Intel on the project.
So far HP has announced an Atom-based server using Intel’s Centerton processor. It suggested it would use cartridges to flexibly upgrade a single server chassis for a wide range of ARM- and Atom-based chips in 2013 and beyond.
On a three-to-five year horizon, HP Labs is working on what it calls "nanostores." The chips combine memristors and logic that could challenge microprocessors in a new era of designs based on novel system architectures and memory hierarchies, said Parthasarathy Ranganathan, an HP Labs researcher in a keynote at the Server Design Summit here.
“We have the opportunity for new building block,” said Ranganathan. “It’s really a 3-D stack amenable to traditional workloads and even more so to new workloads, really changing the game with potentially a hundred-fold increase in performance per watt."
The memristor terminology is probably for PR. Defects, especially charge trapping type, lend themselves to memory effects pretty easily, and nothing is ever defect free, so a lot of these defect-driven memories can appear under the right stress conditions.
Charge trapping models for resistance switching would be open to objections which apply to all models which do not involve stable atomic/ionic rearrangements, i.e., phase changes, in localized regions of the considered materials: the memory states last for several years without decay, and it is hard to see that non-equilibrium charge distributions could exist for so long in thin-film devices.
I believe A.Sceptic's issue is not with "Hynix and other companies" but rather with those who write memristor papers and are either not aware of the physics issues or have made no real attempt to incorporate the actual physics into dynamic systems models (regardless of whether it is called a "memristor" or not).
The lead HP memristor scientist (Stan Williams) has publically stated in an earlier EETimes article that HP's memristor research is "essentially complete" and any delay is for business reasons blamed on Hynix. However, to my knowledge, there is no realistic model from HP or anyone else incorporating the known physics into a dynamic systems model (regardless of whether you call it a memristor or not). The lack of accurate physics models will make product design difficult and Hynix's engineers will likely need to rely on trial and error for manufacture. This will likely make it more costly and time consuming to manufacture "memristors".
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.