The most popular and comprehensive Open Source ECM platform
Emerging Memories: New Technologies Targets AI and IoT
Artificial Technology and Internet of Things (IoT) devices are driving the development of alternative hardware memory technologies. The use of SRAM and DRAM is expensive and consume large amounts of power.
Rob Crooke, senior vice president at Intel, said that “DRAM is not big enough to solve today’s problem of real-time data analysis—and traditional storage isn’t fast enough.”
Emerging memory technologies are typically non-volatile and often much faster than solid state drives. Some of the promising emerging memory technologies include:
Intel Optane DC – Non-volatile, high-capacity memory accessible through memory operations or by blocks. The target use case for Optane is in the data center.
MRAM – Non-volatile memory that can be completely powered down but which can be powered on quickly and which supports fast writes. MRAM fits the IoT use case well and offers greater capacity and density and requires less power.
ReRAM (Resistive Random Access Memory) – Has very low read latency and fast write performance. ReRam is well suited for in-memory computing.
David Fried, chief technology officer at Coventor, now part of Lam Research, said that “I would expect these advanced memories to first find homes in applications that recognize or leverage one of their unique advantages.”
0 comments on “Emerging Memories: New Technologies Targets AI and IoT”
1 Pings/Trackbacks for "Emerging Memories: New Technologies Targets AI and IoT"
-
[…] To Read More, Please Visit Source […]