OPTIMA: Design-Space Exploration of Discharge-Based In-SRAM Computing: Quantifying Energy-Accuracy Trade-offs

Abstract

In-SRAM computing promises energy efficiency, but circuit nonlinearities and PVT variations pose major challenges in designing robust accelerators. To address this, we introduce OPTIMA, a modeling framework that aids in analyzing bitline discharge and power consumption in 6T-SRAM-based accelerators. It provides insights into limiting factors and enables fast design-space exploration of circuit configurations. Leveraging OPTIMA for in-SRAM multiplications exhibits ∼100 × simulation speed-up while maintaining an RMS modeling error of 0.88 mV. Exploration yields an optimized multiplier with 1.05 pJ energy consumption per 4-bit operation and classification accuracies of 71.8% (top-1) and 90.4% (top-5) for ImageNet and 92.5% for CIFAR-10 datasets respectively when applied in quantized DNNs. To further support research and development efforts, we have made our OPTIMAptima tool flow available as open source [https://github.com/sevjaeg/optima].