Exploiting Partially-Forgetful Memories for Approximate Computing

While the memory subsystem is already a major contributor to energy consumption of embedded systems, the guard-banding required for masking the effects of ever increasing manufacturing variations in memories imposes even more energy overhead. In this letter, we explore how partially-forgetful memori...

Full description

Saved in:
Bibliographic Details
Published inIEEE embedded systems letters Vol. 7; no. 1; pp. 19 - 22
Main Authors Shoushtari, Majid, BanaiyanMofrad, Abbas, Dutt, Nikil
Format Journal Article
LanguageEnglish
Published IEEE 01.03.2015
Subjects
Approximate computing
Benchmark testing
Data structures
Hardware
memory hierarchy
process variation
PSNR
Random access memory
Reliability
Software
Online AccessGet full text

Cover

More Information
Summary:While the memory subsystem is already a major contributor to energy consumption of embedded systems, the guard-banding required for masking the effects of ever increasing manufacturing variations in memories imposes even more energy overhead. In this letter, we explore how partially-forgetful memories can be used by exploiting the intrinsic tolerance of a vast class of applications to some level of error for relaxing this guard-banding in memories. We discuss the challenges to be addressed and introduce relaxed cache as an exemplar to address these challenges for partially-forgetful SRAM caches. Preliminary results show how adapting guard-bands to application characteristics can help the system save significant amount of cache leakage energy (up to 74%) while still generating acceptable quality results.
ISSN:1943-0663
1943-0671
DOI:10.1109/LES.2015.2393860