Title
Optimizing job performance under a given power constraint in HPC centers
Abstract
Never-ending striving for performance has resulted in a tremendous increase in power consumption of HPC centers. Power budgeting has become very important from several reasons such as reliability, operating costs and limited power draw due to the existing infrastructure. In this paper we propose a power budget guided job scheduling policy that maximize overall job performance for a given power budget. We have shown that using DVFS under a power constraint performance can be significantly improved as it allows more jobs to run simultaneously leading to shorter wait times. Aggressiveness of frequency scaling applied to a job depends on instantaneous power consumption and on the job's predicted performance. Our policy has been evaluated for four workload traces from systems in production use with up to 4 008 processors. The results show that our policy achieves up to two times better performance compared to power budgeting without DVFS. Moreover it leads to 23% lower CPU energy consumption on average. Furthermore, we have investigated how much job performance and energy efficiency can be improved under our policy and same power budget by an increase in the number of DVFS enabled processors.
Year
DOI
Venue
2010
10.1109/GREENCOMP.2010.5598303
Green Computing Conference
Keywords
Field
DocType
limited power,optimizing job performance,power budgeting,job scheduling policy,hpc center,power constraint performance,overall job performance,job performance,power consumption,instantaneous power consumption,power budget,times better performance,frequency scaling,transistors,energy efficient,job scheduling
Power budget,Computer science,Efficient energy use,Workload,Real-time computing,Frequency scaling,Job scheduler,Power (physics),Job performance,Energy consumption,Reliability engineering
Conference
Citations 
PageRank 
References 
28
1.20
22
Authors
4
Name
Order
Citations
PageRank
Maja Etinski11215.67
Julita Corbalan217412.00
Jesus Labarta3138992.33
Mateo Valero44520355.94