LONDON – France's Centre National de la Recherche Scientifique (CNRS) has done preliminary work on several nanoelectronic memory technologies that it now wishes to progress further through licensing agreements.
The barriers and potential for the voltage- and density-scaling of established memory devices – including DRAM, SRAM, Flash and phase-change memory – beyond the 22-nm node is one of hot topics for current memory R&D.
The technologies developed by CNRS include the relatively broadly researched resistive RAM and three rather more obscure technologies that make use of electrical and magnetic effects in physical structures likely to exist below 22-nm. CNRS, which is funded by the French government, has filed patents applications for all of these memories.
While it is not certain which, if any, of these technologies will become significant in the semiconductor industry, business opportunities come from science and engineering breakthroughs which in turn come from academic research. CNRS has six laboratories working on novel memories.
France Innovation et Scientifique Transfert SA (FIST; Paris, France) is responsible for the licensing and the transfer of technologies from CNRS to the commercial sector.
I suppose this could be a case where some of the "big boys" developing these technologies would be willing to throw some bread crumbs (relatively speaking) to perhaps get an idea or two. Perhaps the limit of the French govt. funding was to file patent disclosures...
What does this mean? Has CNRS run out of money to take these technologies further? is it simply a way to monetise prior investment? or is there a real win-win opportunity in CNRS' offer?
What do you think, EEtimers?
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.