Search
Exact matches only
Search in title
Search in content
Filter by Post Type
Categories
Conference (i)
Journal (ii)
Book (iii)
Graduate Work (iv)
Patent (v)
Magazine (vi)
Movie (vii)
Category (viii)
Others (ix)
Identifier
Country Code
Filter by Presenters
Filter by Keywords
Content from
Content to

SwiftGPU: Fostering Energy Efficiency in a Near-Threshold GPU through a Tactical Performance Boost.

Presenters

Contributors


Description

"In this paper, we investigate the challenges of preserving energy-efficiency in a Near-Threshold Computing (NTC) GPU. Two key factors can significantly undermine the efficacy of GPUs at NTC: (a) elongated delays at NTC make the GPU applications severely sensitive toMulti-cycle Latency Datapaths (MLDs) within the GPU pipeline; and (b) process variation (PV) at NTC induces a substantial performance variance. To address these emerging challenges, we propose SwiftGPU---an energyefficient GPU design paradigm at NTC. SwiftGPU dynamically adjusts the degree of parallelization, and the speed of the MLDs within each stream core of the GPU. The proposed scheme achieves an average of~15% improvement in energy-efficiency over an ideal PV-free GPU, operating at the Super-Threshold regime. SwiftGPU incurs marginal area, wire-length and power overheads of 0.65%, 2.6% and 3.7%, respectively."

Categories

Conference (i)

Keywords

Country Code

US

Year

2018

Month

4

Identifier

2018.4.i.000000471.US.0

Submit a Comment

Comment Submission Guidelines
Intuitive and thought-provoking comments pertaining to the content are welcome for discussion and any misconduct in the form of posting off-topic or promotional content might not be given the permission to be published as per the moderators’ discernment. All postings come under the authority of Abstract Tube Inc. *