Extending the task scheduling of the LAM/MPI library - DOI: 10.4025/actascitechnol.v30i2.5461
DOI:
https://doi.org/10.4025/actascitechnol.v30i2.5461Keywords:
task scheduling, distributed applications, LAM/MPIAbstract
Computer cluster performance is directly related to how the workload is distributed among its nodes. Task scheduling techniques are employed to optimize the usage of system resources, thus improving the performance of parallel applications. This work presents an extension to the LAM/MPI library that allows process scheduling to be done more efficiently than in its original implementation. Tests performed with three MPI applications, running on an 8-node cluster, show that our solutions can reduce execution time by over 50% in most cases.Downloads
Download data is not yet available.
Downloads
Published
2008-10-20
Issue
Section
Computer Science
License
DECLARATION OF ORIGINALITY AND COPYRIGHTS
I Declare that current article is original and has not been submitted for publication, in part or in whole, to any other national or international journal.
The copyrights belong exclusively to the authors. Published content is licensed under Creative Commons Attribution 4.0 (CC BY 4.0) guidelines, which allows sharing (copy and distribution of the material in any medium or format) and adaptation (remix, transform, and build upon the material) for any purpose, even commercially, under the terms of attribution.
Read this link for further information on how to use CC BY 4.0 properly.
How to Cite
Valle Rego Gorino, F., Martini, J. A., & de Lara Gonçalves, R. A. (2008). Extending the task scheduling of the LAM/MPI library - DOI: 10.4025/actascitechnol.v30i2.5461. Acta Scientiarum. Technology, 30(2), 125-133. https://doi.org/10.4025/actascitechnol.v30i2.5461
8.png){kind=logo}
