SCiMA: A Generic Single-Cycle Compute-in-Memory Acceleration Scheme for Matrix Computations

This work proposes a new generic Single-cycle Compute-in-Memory (CiM) Accelerator for matrix computation named SCiMA. SCiMA is developed on top of the existing commodity Spin-Orbit Torque Magnetic Random-Access Memory chip. Every sub-array's peripherals are transformed to realize a full set of single-cycle 2- and 3-input in-memory bulk bitwise functions specifically designed to accelerate a wide variety of graph and matrix multiplication tasks. We explore SCiMA's efficiency by selecting a complex matrix processing operation, i.e., calculating determinant as an essential and under-explored application in the CiM domain. The cross-layer device-to-architecture simulation framework shows the presented platform can reduce energy consumption by 70.43% compared with the most recent CiM designs implemented with the same memory technology. SCiMA also achieves up to 2.5x speedup compared with current CiM platforms.

Automated summary

This work proposes a new generic Single-cycle Compute-in-Memory (CiM) Accelerator named SCiMA for matrix computation. By utilizing specifically designed single-cycle in-memory bulk bitwise functions, SCiMA can accelerate a wide variety of graph and matrix multiplication tasks. Experimental results show that SCiMA can reduce energy consumption by 70.43% compared with the most recent CiM designs implemented with the same memory technology and achieve up to 2.5x speedup compared with current CiM platforms.