Logo image
A 48 TOPS and 20943 TOPS/W 512kb Computation-in-SRAM Macro for Highly Reconfigurable Ternary CNN Acceleration
Conference paper

A 48 TOPS and 20943 TOPS/W 512kb Computation-in-SRAM Macro for Highly Reconfigurable Ternary CNN Acceleration

Chih-Sheng Lin, Fu-Cheng Tsai, Jian-Wei Su, Sih-Han Li, Tian-Sheuan Chang, Shyh-Shyuan Sheu, Wei-Chung Lo, Shih-Chieh Chang, Chih-I Wu and Tuo-Hung Hou
Proceedings - A-SSCC 2021: IEEE Asian Solid-State Circuits Conference
2021

Abstract

Hardware and Architecture Signal Processing Electrical and Electronic Engineering
Energy-and area-efficient acceleration solutions are critical for the continual development of ubiquitous, real-Time, and cross-domain artificial intelligence [1]-[7]. This motivates the active research on the SRAM-based computing-in-memory (CIM) that exploits the state-of-The-Art CMOS technology and massively parallel analog computing directly inside the memory array [1]-[4]. Although significant progress has been made in recent years in improving throughput [1-2], energy efficiency [1], [3], and area efficiency [1], [4], simultaneously achieving them in SRAM-CIM remains an unsolved problem. This is particularly challenging when accounting for the potential accuracy loss due to nonideality in analog computing and the inflexibility of CIM weight-stationary designs.

Metrics

1 Record Views

Details

Logo image