July 10, 2007 — Grandis, Inc. has received a Small Business Innovation Research (SBIR) Phase I grant from the National Science Foundation (NSF) to develop and commercialize spin-transfer torque random-access memory (STT-RAM). STT-RAM is a next-generation non-volatile technology designed to enable excellent scalability and unlimited endurance with low power and fast read/write capability. The NSF grant will support Grandis’ research into cell architectures that enhance the thermal stability of STT-RAM.

According to Grandis co-founder and CTO Yiming Huai, Ph.D., “Today’s conventional memory technologies face severe scaling and performance challenges at or below the 45-nm node. We believe STT-RAM can fill this gap as the first truly universal and scalable ultra-fast, non-volatile memory solution.”

To date, the NSF has awarded Grandis approximately $700,000 in grants to develop its STT-RAM technology, which exploits the spin of electrons for its operation. Its inherent low 1.2-volt internal voltage is in stark contrast to current technologies such as DRAM and Flash. Grandis says that its performance, particularly its sub-10-nanosecond write time and unlimited endurance, also exceeds that of other prospective non-volatile memory technologies, such as phase-change RAM (PRAM). And its low writing current, which can continue to scale down with shrinking design rules, promises greater density and, ultimately, lower cost per die.