DocumentCode
3024253
Title
Meeting Deadlines Cheaply
Author
Legriel, Julien ; Maler, Oded
Author_Institution
CNRS, Gieres, France
fYear
2011
fDate
5-8 July 2011
Firstpage
185
Lastpage
194
Abstract
We develop a computational framework for solving the problem of finding the cheapest configuration (in terms of the number of processors and their respective speeds) of a multiprocessor architecture on which a task graph can be scheduled within a given deadline. We then extend the problem in three orthogonal directions: taking communication volume into account, considering the case where a stream of instances of the task graph arrives periodically and reformulating the problem as a bi-criteria optimization for which we approximate the Pareto front.
Keywords
Pareto optimisation; graph theory; multiprocessing systems; processor scheduling; task analysis; Pareto front; bicriteria optimization; cheapest configuration; computational framework; multiprocessor architecture; scheduling; task graph; Computer architecture; Encoding; Optimal scheduling; Processor scheduling; Program processors; Schedules;
fLanguage
English
Publisher
ieee
Conference_Titel
Real-Time Systems (ECRTS), 2011 23rd Euromicro Conference on
Conference_Location
Porto
ISSN
1068-3070
Print_ISBN
978-1-4577-0643-1
Type
conf
DOI
10.1109/ECRTS.2011.25
Filename
6001780
Link To Document