SEGA-DCIM: Design Space Exploration-Guided Automatic Digital CIM Compiler with Multiple Precision Support

TL;DR

SEGA-DCIM is an automated compiler for digital computing-in-memory designs that uses design space exploration and multi-precision support to optimize area, power, and delay, reducing manual effort and design time.

Contribution

It introduces an automatic DCIM compiler with multi-precision support and a genetic algorithm-based design space exploration method.

Findings

Provides a wide design space including integer and floating-point designs.

Achieves competitive performance with state-of-the-art DCIMs.

Automates netlist and layout generation for DCIMs.

Abstract

Digital computing-in-memory (DCIM) has been a popular solution for addressing the memory wall problem in recent years. However, the DCIM design still heavily relies on manual efforts, and the optimization of DCIM is often based on human experience. These disadvantages limit the time to market while increasing the design difficulty of DCIMs. This work proposes a design space exploration-guided automatic DCIM compiler (SEGA-DCIM) with multiple precision support, including integer and floating-point data precision operations. SEGA-DCIM can automatically generate netlists and layouts of DCIM designs by leveraging a template-based method. With a multi-objective genetic algorithm (MOGA)-based design space explorer, SEGA-DCIM can easily select appropriate DCIM designs for a specific application considering the trade-offs among area, power, and delay. As demonstrated by the experimental results, SEGA-DCIM offers solutions with wide design space, including integer and floating-point precision designs, while maintaining competitive performance compared to state-of-the-art (SOTA) DCIMs.