Answers are grounded in this paper and cite their sources.
Computing-in-memory (CIM) is an emerging approach for alleviating the Von-Neumann bottleneck of latency and energy overheads, and improving energy efficiency and throughput. In this brief, we present a novel CIM macro aimed at improving the energy efficiency and throughput of edge devices when running 4b multiply-and-accumulate (MAC) operations. The proposed architecture uses (1) a customized 9T1C bit-cell in charge-domain computation for sensing margin improvement and compact design; (2) a hierarchical capacity attenuator for 4b weight accumulation without complicated controlling switches and signals for throughput improvement; (3) an input sparsity-sensing-based flash analog-to-digital converters readout scheme to improve energy efficiency and throughput. Fabricated in 28nm CMOS technology, the proposed 32Kb SRAM CIM macro demonstrates an average energy efficiency of 646.6 TOPS/W (normalized to 4b/4b input/weight) and a throughput of 1638.4 GOPS while achieving 84.89% classification accuracy on the CIFAR-10 dataset at 4b precision in inputs and weights.
No takes yet. Share an insight, caveat, or question.
Xiao et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: