Warp: A Hardware Platform for Efficient Multi- Modal Sensing with Adaptive Approximation

Authors
Stanley-Marbell, Phillip  ORCID logo  https://orcid.org/0000-0001-7752-2083
Rinard, Martin 
DOI
https://doi.org/10.1109/MM.2019.2951004.
Thumbnail Image
Files
Accepted version (3.43 MB)
Type
Article
Change log
Abstract

We present Warp, the first open hardware platform designed explicitly to support research in approximate computing. Warp incorporates 21 sensors, computation, and circuit-level facilities designed explicitly to enable approximate computing research, in a 3.6 cm×3.3 cm×0.5 cm area. Warp uses these facilities to support a wide range of precision and accuracy versus power and performance tradeoffs.

Publication Date
Online Publication Date
2020-01-14
Acceptance Date
2018-09-19
Keywords
Journal Title
IEEE Micro
Journal ISSN
0272-1732
Volume Title
40
Publisher
IEEE
Rights
http://www.rioxx.net/licenses/all-rights-reserved
Sponsorship
Royal Society (RG170136)
Engineering and Physical Sciences Research Council (EP/R022534/1)
Engineering and Physical Sciences Research Council (EP/P001246/1)
Alan Turing Institute (EP/N510129/1)
Collections
Cambridge University Research Outputs