Frontiers of Information Technology & Electronic Engineering
>> 2020,
Volume 21,
Issue 10
doi:
10.1631/FITEE.2000089
Multi-dimensional optimization for approximate near-threshold computing
Affiliation(s): College of Information Engineering, Capital Normal University, Beijing 100056, China; Institution of Spacecraft System Engineering China Academy of Space Technology, Beijing 100094, China; Beijing Advanced Innovation Center for Imaging Theory and Technology, Beijing 100048, China; less
Received: 2020-02-26
Accepted: 2020-10-14
Available online: 2020-10-14
Next
Previous
Abstract
The demise of Dennard’s scaling has created both power and utilization wall challenges for computer systems. As transistors operating in the near-threshold region are able to obtain flexible trade-offs between power and , it is regarded as an alternative solution to the scaling challenge. A reduction in supply voltage will nevertheless generate significant reliability challenges, while maintaining an error-free system that generates high costs in both and consumption. The main purpose of research on computer architecture has therefore shifted from improvement to complex multi-objective optimization. In this paper, we propose a three-dimensional optimization approach which can effectively identify the best system configuration to establish a balance among , , and reliability. We use a dynamic programming algorithm to determine the proper voltage and approximate level based on three predictors: system , consumption, and output quality. We propose an which uses a hardware/software co-design fault injection platform to evaluate the impact of the error on output quality under (NTC). Evaluation results demonstrate that our approach can lead to a 28% improvement in output quality with a 10% drop in overall efficiency; this translates to an approximately 20% average improvement in accuracy, power, and .
京公网安备 11010502051620号
