|
For Full-Text PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
|
Scheduling Real-Time Multi-Processor Systems with Distance-Constrained Tasks Using the Early-Release-Fair Model
Da-Ren CHEN Chiun-Chieh HSU Chien-Min WANG
Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Vol.E89-A
No.11
pp.3260-3271 Publication Date: 2006/11/01 Online ISSN: 1745-1337
DOI: 10.1093/ietfec/e89-a.11.3260 Print ISSN: 0916-8508 Type of Manuscript: PAPER Category: Digital Signal Processing Keyword: distance-constrained task systems, real-time scheduling, Pfair, ERfair, rate-monotonic,
Full Text: PDF(1.2MB)>>
Summary:
A hard real-time system is one whose correctness depends not only on the logical result, but also when the results are produced. While many techniques have been proposed for single processor real-time systems, multiprocessor systems have not been studied so extensively. In this paper, we mainly propose two variant (DCTS) by using the Early-Release-Fair (ERfair) and Proportionate-fair (Pfair) model with integral assumptions for identical multi-processor real-time systems. ERfair is a scheduling model for real-time tasks on a multiprocessor system. On the different definitions of distance constraint, we propose two efficient scheduling algorithms designed to probe whether the distance constraints of all ER-fair tasks can be guaranteed. If the distance constraints cannot be guaranteed, then the proposed algorithms gather the unfeasible tasks and inflate them with a reweighting function. The proposed algorithms are linear-time and most suitable for dynamic systems. The experimental results reveal that the proposed algorithms increase significantly the ratio of schedulable task sets.
|
open access publishing via
|
|
|
|
|
|
|
|