Power-performance optimization of a virtualized SMT vector processor via thread fusion and lane configuration

Yaojie Lu, Seyedamin Rooholamin, Sotirios G. Ziavras

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Lane-based Vector Processors (VPs) are highly scalable. However, they become less energy efficient as they scale for vector applications having insufficient data-level parallelism (DLP) to keep the extra computation lanes fully occupied. We present a scalable and yet flexible VP that is capable of dynamically deactivating some of its computing lanes in order to reduce static power with minimum performance loss. In addition, our simultaneous multi-threaded (SMT) VP can exploit identical instruction flows that may be present in different vector applications by running in a novel fused mode that increases its utilization. We introduce a power model and two optimization policies for minimizing the consumed energy, or the product of the energy and runtime for a given application. Benchmarking that involves an FPGA prototype shows up to 33.8% energy reduction in addition to 40% runtime improvement, or up to 62.7% reduction in the product of energy and runtime.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2016
PublisherIEEE Computer Society
Pages81-86
Number of pages6
ISBN (Electronic)9781467390385
DOIs
StatePublished - Sep 2 2016
Event15th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2016 - Pittsburgh, United States
Duration: Jul 11 2016Jul 13 2016

Publication series

NameProceedings of IEEE Computer Society Annual Symposium on VLSI, ISVLSI
Volume2016-September
ISSN (Print)2159-3469
ISSN (Electronic)2159-3477

Other

Other15th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2016
Country/TerritoryUnited States
CityPittsburgh
Period7/11/167/13/16

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Hardware and Architecture

Keywords

  • DLP
  • dynamic configuration
  • instruction fusion
  • performance and energy optimization
  • vector processor

Fingerprint

Dive into the research topics of 'Power-performance optimization of a virtualized SMT vector processor via thread fusion and lane configuration'. Together they form a unique fingerprint.

Cite this