A Resilient Interface for Approximate Data Access

Full text
Author(s):	Fabricio Filho, Joao ; Felzmann, Isaias B. ; Azevedo, Rodolfo J. ; Wanner, Lucas F. ; IEEE Total Authors: 5
Document type:	Journal article
Source:	2022 25TH INTERNATIONAL CONFERENCE ON INFORMATION FUSION (FUSION 2022); v. N/A, p. 8-pg., 2019-01-01.
Abstract
Approximate memories offer lower energy cost while introducing errors to applications. If such errors affect essential parts of the application, the execution may fail, decreasing outputs quality and energy savings. We present AxRAM, a memory architecture interface for approximate data that allows applications to benefit from energy savings provided by approximate memories, while improving quality of results and failure rates. AxRAM places critical application data into non-approximate memory regions and implements a resilient addressing scheme that reduces invalid data accesses that lead to application crashes. On a voltage-overscaled SRAM, AxRAM reduces 50.97% of application crashes on average when compared to an unprotected approximate memory. Compared to a system using non approximate SRAM, our proposal reduces energy consumption by 4536% on average while providing at least 80% output quality. (AU)

FAPESP's process:	18/24177-0 - Architectural Support for Approximate Computing
Grantee:	Isaías Bittencourt Felzmann
Support Opportunities:	Scholarships in Brazil - Doctorate