A Calibratable Model for Fast Energy Estimation of MVM Operations on RRAM Crossbars

TL;DR

This paper presents a calibratable model for rapid and accurate energy estimation of matrix-vector multiplication operations on RRAM crossbars, aiding the design of energy-efficient AI hardware.

Contribution

It introduces a novel abstract model and calibration method that simplifies circuit-level energy estimation for RRAM-based CIM architectures.

Findings

Speedup of up to 1000x in energy estimation

Estimation errors below 1% compared to SPICE simulations

Effective tool for DNN workload energy assessment

Abstract

The surge in AI usage demands innovative power reduction strategies. Novel Compute-in-Memory (CIM) architectures, leveraging advanced memory technologies, hold the potential for significantly lowering energy consumption by integrating storage with parallel Matrix-Vector-Multiplications (MVMs). This study addresses the 1T1R RRAM crossbar, a core component in numerous CIM architectures. We introduce an abstract model and a calibration methodology for estimating operational energy. Our tool condenses circuit-level behaviour into a few parameters, facilitating energy assessments for DNN workloads. Validation against low-level SPICE simulations demonstrates speedups of up to 1000x and energy estimations with errors below 1%.