2010 IEEE International Symposium on Parallel &Amp; Distributed Processing, Workshops and PHD Forum (IPDPSW) 2010
DOI: 10.1109/ipdpsw.2010.5470913
|View full text |Cite
|
Sign up to set email alerts
|

BSLD threshold driven power management policy for HPC centers

Abstract: In this paper, we propose a power-aware parallel job scheduler assuming DVFS enabled clusters. A CPU frequency assignment algorithm is integrated into the well established EASY backfilling job scheduling policy. Running a job at lower frequency results in a reduction in power dissipation and accordingly in energy consumption. However, lower frequencies introduce a penalty in performance. Our frequency assignment algorithm has two adjustable parameters in order to enable fine grain energy-performance trade-off … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2015
2015
2017
2017

Publication Types

Select...
3

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(2 citation statements)
references
References 28 publications
0
2
0
Order By: Relevance
“…The degradation has been measured for the NAS benchmark [9], the SPEC float and integer benchmarks. A degradation of 163% is assumed to be a good approximation [26]. Table III In our context, the SHUT policy appears to be the best one.…”
Section: A Presentation Of Curiementioning
confidence: 94%
“…The degradation has been measured for the NAS benchmark [9], the SPEC float and integer benchmarks. A degradation of 163% is assumed to be a good approximation [26]. Table III In our context, the SHUT policy appears to be the best one.…”
Section: A Presentation Of Curiementioning
confidence: 94%
“…The first policy exploits energy savings that can be achieved by running jobs at reduced frequency when the system load is low [14]. The other policy decides whether to run a job at reduced frequency based on job performance prediction [12]. This part of the thesis is aimed at discovering real potentials of DVFS focusing on the entire workload.…”
Section: Contributionsmentioning
confidence: 99%