DocumentCode
3139503
Title
A Hybrid Task Scheduling for Multi-Core Platform
Author
Lee, Liang-Teh ; Chang, Huang-Yuan ; Chao, Shu-Wei
Author_Institution
Dept. of Comput. Sci. & Eng., Tatung Univ.
Volume
5
fYear
2008
fDate
13-15 Dec. 2008
Firstpage
40
Lastpage
45
Abstract
With the advancement of technology, various services can be combined in a single computer system, such as Mobile Computing, and Digital Home, etc. With the properties of compact size, low power consumption, and full computing power, multi-core processor systems can be used to fulfill these kind of applications. For a set of hybrid tasks which comprise large real-time and non-real-time applications, however, currently, the operating system does not have any scheduling policy to improve the performance on multi-core platforms effectively. The task scheduling of operating system for general purpose applications can not satisfy the working demand of hybrid tasks, especially with large amount of real-time tasks. In this paper, we proposed a hybrid task scheduling scheme for multi-core platforms. In the proposed scheme, a two-level hierarchical scheduling is applied to adjust real-time and non-real-time tasks. It can not only maintain the response time of general tasks, but also support the real-time requirements for other tasks. The experimental results show that a higher efficiency can be obtained by applying the proposed scheme on multi-core architectures.
Keywords
processor scheduling; digital home; hybrid task scheduling; mobile computing; multicore platform; multicore processor systems; operating system; real-time requirements; two-level hierarchical scheduling; Application software; Delay; Home computing; Multicore processing; Operating systems; Power engineering computing; Processor scheduling; Real time systems; Runtime; Scheduling algorithm;
fLanguage
English
Publisher
ieee
Conference_Titel
Future Generation Communication and Networking Symposia, 2008. FGCNS '08. Second International Conference on
Conference_Location
Sanya
Print_ISBN
978-1-4244-3430-5
Electronic_ISBN
978-0-7695-3546-3
Type
conf
DOI
10.1109/FGCNS.2008.152
Filename
4813630
Link To Document