Lower-bound complexity algorithm for task scheduling on heterogeneous grid

The problem of best schedule of dependent-tasks application into nodes of computational grid in low complexity is the most important issue to obtain high performance application execution. Scheduling can be performed at compile-time or at run-time depends on tasks and grid-nodes available informatio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computing Jg. 99; H. 11; S. 1125 - 1145
Hauptverfasser: Atef, Asmaa, Hagras, Tarek, Mahdy, Yousef B., Janeček, Jan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Vienna Springer Vienna 01.11.2017
Springer Nature B.V
Schlagworte:
Algorithms
Artificial Intelligence
Computer Appl. in Administrative Data Processing
Computer Communication Networks
Computer Science
Computer simulation
Heuristic
Information Systems Applications (incl.Internet)
Reproduction (copying)
Scheduling
Scheduling algorithms
Software Engineering
Task complexity
Task scheduling
Task scheduling
Grid computing
Heterogeneous grids
68M14
ISSN:0010-485X, 1436-5057
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The problem of best schedule of dependent-tasks application into nodes of computational grid in low complexity is the most important issue to obtain high performance application execution. Scheduling can be performed at compile-time or at run-time depends on tasks and grid-nodes available information. Due to the NP-completeness of the problem, heuristics are used in compile-time solution of the problem. Two of these heuristics are list-scheduling and duplication-based. List-scheduling heuristics produce reasonable schedule in reasonable time complexity, while duplication-based heuristics produce better schedule in higher time complexity. Many algorithms based on list-scheduling and duplication-based heuristics have been addressed in the literature. This paper proposes a scheduling algorithm based on list-scheduling and duplication-based heuristics. The algorithm is called Best-Node based Critical-Parent. The algorithm keeps the lower-bound complexity of any classes of heuristics-based scheduling algorithms. Random generated applications, in addition to real-world applications have been examined. The experimental results based on computer simulation show that the proposed algorithm performed better than the most-recent and well-known existing algorithms.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0010-485X
1436-5057
DOI:10.1007/s00607-017-0558-5