Optimizing runtime reconfiguration decisions

Thilo Pionteck, Steffen Sammann, Carsten Albrecht

Abstract

Partially reconfigurable hardware accelerators enable the offloading of computative intensive tasks from software to hardware at runtime. Beside handling the technical aspects, finding a proper reconfiguration point in time is of great importance for the overall system performance. Determination of a suitable point of reconfiguration demands the evaluation of performance degradation during runtime reconfiguration and expected performance benefit after reconfiguration. Three different approaches to determine a proper point of reconfiguration are discussed. Delays and weighted transitions are used to reduce the number of reconfigurations while keeping system performance at a maximum. Evaluation is done with a simulation model of a runtime reconfigurable network coprocessor. Results show that the number of reconfigurations can be reduced by about 35% for a given application scenario. By optimizing runtime reconfiguration decisions, the overall system performance is even higher than compared to pure threshold based reconfiguration decision schemes.

Original languageEnglish
Title of host publication2010 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing
Number of pages8
PublisherIEEE
Publication date01.12.2010
Pages39-46
Article number5703496
ISBN (Print)978-1-4244-9719-5
ISBN (Electronic)978-0-7695-4322-2
DOIs
Publication statusPublished - 01.12.2010
EventIEEE/IFIP 8th International Conference on Embedded and Ubiquitous Computing - Hong Kong, Hong Kong
Duration: 11.12.201013.12.2010
Conference number: 83896

Fingerprint

Dive into the research topics of 'Optimizing runtime reconfiguration decisions'. Together they form a unique fingerprint.

Cite this