A Fuzzy algorithm for parallelizability evaluation and load balance on the multi-core processor

Author

Dai Hongjun ; Xu XinShun ; Jia Zhiping

Author_Institution

Dept. of Comput. Sci. & Technol., Shandong Univ., Jinan

fYear

2008

fDate

26-28 Aug. 2008

Firstpage

168

Lastpage

171

Abstract

To schedule real-time threads on the multi-core processors, a fuzzy algorithm is presented for parallelizability evaluation and load balance, which can be used into the operating system scheduling or the hardware controller design. It is based on the fuzzy inference system (FIS) to derive answers from a knowledge base with a fuzzy inference engine and the rules are taken deadline, laxity and interval into account. The experiments and results show that the algorithm can improve the performance of the overall multi-core threading execution.

Keywords

fuzzy set theory; inference mechanisms; knowledge based systems; multi-threading; processor scheduling; resource allocation; fuzzy inference system; hardware controller design; knowledge base; load balancing; multicore processor; multicore threading execution; operating system scheduling; parallelizability evaluation; real-time thread scheduling; Fuzzy control; Fuzzy systems; Hardware; Inference algorithms; Multicore processing; Operating systems; Processor scheduling; Real time systems; Scheduling algorithm; Yarn;

fLanguage

English

Publisher

ieee

Conference_Titel

Granular Computing, 2008. GrC 2008. IEEE International Conference on

Conference_Location

Hangzhou

Print_ISBN

978-1-4244-2512-9

Electronic_ISBN

978-1-4244-2513-6

Type

conf

DOI

10.1109/GRC.2008.4664678

Filename

4664678