Utility-based resource management in an oversubscribed energy-constrained heterogeneous environment executing parallel applications
Name:
parco_manuscript.pdf
Size:
1.013Mb
Format:
PDF
Description:
Final Accepted Manuscript
Author
Machovec, DylanKhemka, Bhavesh
Kumbhare, Nirmal
Pasricha, Sudeep
Maciejewski, Anthony A.
Siegel, Howard Jay
Akoglu, Ali
Koenig, Gregory A.
Hariri, Salim
Tunc, Cihan
Wright, Michael
Hilton, Marcia
Rambharos, Rajendra
Blandin, Christopher
Fargo, Farah
Louri, Ahmed
Imam, Neena
Affiliation
Univ Arizona, Dept Elect & Comp EngnIssue Date
2019-04Keywords
Heterogeneous computingEnergy-aware computing
Utility functions
Resource management heuristics
Parallel tasks
Scheduling
Metadata
Show full item recordPublisher
ELSEVIER SCIENCE BVCitation
Machovec, D., Khemka, B., Kumbhare, N., Pasricha, S., Maciejewski, A. A., Siegel, H. J., ... & Wright, M. (2017). Utility-based resource management in an oversubscribed energy-constrained heterogeneous environment executing parallel applications. Parallel Computing.Journal
PARALLEL COMPUTINGRights
© 2017 Elsevier B.V. All rights reserved.Collection Information
This item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at repository@u.library.arizona.edu.Abstract
The worth of completing parallel tasks is modeled using utility functions, which monotonically-decrease with time and represent the importance and urgency of a task. These functions define the utility earned by a task at the time of its completion. The performance of a computing system is measured as the total utility earned by all completed tasks over some interval of time (e.g., 24 h). We have designed, analyzed, and compared the performance of a set of heuristic techniques to maximize system performance when scheduling dynamically arriving parallel tasks onto a high performance computing (HPC) system that is oversubscribed and energy constrained. We consider six utility-aware heuristics and four existing heuristics for comparison. A new concept of temporary place holders is compared with scheduling using permanent reservations. We also present a novel energy filtering technique that constrains the maximum energy-per-resource used by each task. We conducted a simulation study to evaluate the performance of these heuristics and techniques in multiple energy-constrained oversubscribed HPC environments. We conduct an experiment with a subset of the heuristics on a physical testbed system for one scheduling scenario. We demonstrate that our proposed utility-aware resource management heuristics are able to significantly outperform existing techniques. (C) 2017 Elsevier B.V. All rights reserved.Note
24 month embargo; published online: 7 November 2017ISSN
01678191Version
Final accepted manuscriptSponsors
U.S. Department of Energy [DE-AC05-00OR22725]; Department of Energy; Oak Ridge National Laboratory for the Department of Defense (DoD) [4000108022]; National Science Foundation (NSF) [CCF-1302693, CCF-1547036, CCF-1547035]; NSF [CNS-0923386]Additional Links
https://linkinghub.elsevier.com/retrieve/pii/S0167819117301862ae974a485f413a2113503eed53cd6c53
10.1016/j.parco.2017.11.005