ZEM: Zero-cycle Bit-masking Module for Deep Learning Refresh-less DRAM
TimeTuesday, December 7th6:00pm - 7:00pm PST
LocationLevel 2 - Lobby
Event Type
Networking Reception
Work-in-Progress Poster
Virtual Programs
Presented In-Person
DescriptionIn sub-20nm technologies, DRAM cells suffer from poor retention time. With the technology scaling, this problem tends to be worse, significantly increasing refresh power of DRAM. This is more problematic in memory heavy applications such as deep learning systems. In this work, we present a zero-cycle bit-masking(ZEM) scheme that exploits the asymmetry of retention failures, to eliminate DRAM refresh in the inference of CNNs, NLP, and the image generation based on GAN. Our results on 16Gb devices show that ZEM can improve the performance by up to 17.31% while reducing the total energy consumed by DRAM by up to 43.03%