Přístupnostní navigace
E-application
Search Search Close
Publication detail
JAROŠ, M. JAROŠ, J.
Original Title
Performance-Cost Optimization of Moldable Scientific Workflows
Type
conference paper
Language
English
Original Abstract
Moldable scientific workflows represent a special class of scientific workflows where the tasks are written as distributed programs being able to exploit various amounts of computer resources. However, current cluster job schedulers require the user to specify the amount of resources per task manually. This often leads to suboptimal execution time and related cost of the whole workflow execution since many users have only limited experience and knowledge of the parallel efficiency and scaling. This paper proposes several mechanisms to automatically optimize the execution parameters of moldable workflows using genetic algorithms. The paper introduces a local optimization of workflow tasks, a global optimization of the workflow on systems with on-demand resource allocation, and a global optimization for systems with static resource allocation. Several objectives including the workflow makespan, computational cost and the percentage of idling nodes are investigated together with a trade-off parameter putting stress on one objective or another. The paper also discusses the structure and quality of several evolved workflow schedules and the possible reduction in makespan or cost. Finally, the computational requirements of evolutionary process together with the recommended genetic algorithm settings are investigated. The most complex workflows may be evolved in less than two minutes using the global optimization while in only 14s using the local optimization.
Keywords
task graph scheduling, workflow, genetic algorithm, moldable tasks, makespan estimation
Authors
JAROŠ, M.; JAROŠ, J.
Released
30. 9. 2021
Publisher
Springer International Publishing
Location
Portland, Oregon USA
ISBN
978-3-030-88223-5
Book
Job Scheduling Strategies for Parallel Processing
Edition
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
0302-9743
Periodical
Lecture Notes in Computer Science
Number
12985
State
Federal Republic of Germany
Pages from
149
Pages to
167
Pages count
19
URL
https://link.springer.com/book/10.1007%2F978-3-030-88224-2
BibTex
@inproceedings{BUT175770, author="Marta {Jaroš} and Jiří {Jaroš}", title="Performance-Cost Optimization of Moldable Scientific Workflows", booktitle="Job Scheduling Strategies for Parallel Processing", year="2021", series="Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)", journal="Lecture Notes in Computer Science", number="12985", pages="149--167", publisher="Springer International Publishing", address="Portland, Oregon USA", doi="10.1007/978-3-030-88224-2\{_}8", isbn="978-3-030-88223-5", issn="0302-9743", url="https://link.springer.com/book/10.1007%2F978-3-030-88224-2" }