Title :
Job Scheduler Parameter Analysis to Find an Optimal Parameter Configuration
Author_Institution :
Inst. of Phys. & Chem. Res. (RIKEN), Wako
Abstract :
The use of PC cluster systems composed of many PCs is largely spread in these days. It is important to improve system usage keeping proper fair-share policy. This optimization problem is difficult to solve. In scheduler software, there are many parameters and the effectiveness of many parameters is related with each other complicatedly. Therefore it is more difficult to find optimal parameter configuration to improve the system usage efficiency. So in this paper, we performed various analyses of those parameters and report the results.
Keywords :
optimisation; scheduling; systems analysis; PC cluster systems; fair-share policy; job scheduler parameter analysis; optimal parameter configuration; scheduler software; system usage efficiency; Chemical analysis; Competitive intelligence; Grid computing; Indium tin oxide; Performance analysis; Personal communication networks; Resource management; Scheduling algorithm; Software systems; Supercomputers; NQS; PC cluster; Parameter configuration;
Conference_Titel :
Complex, Intelligent and Software Intensive Systems, 2009. CISIS '09. International Conference on
Conference_Location :
Fukuoka
Print_ISBN :
978-1-4244-3569-2
Electronic_ISBN :
978-0-7695-3575-3
DOI :
10.1109/CISIS.2009.203