Exploiting Workload Parallelism for Performance and Power Optimization in Blue Gene

Author

Salapura, Valentina ; Walkup, Robert ; Gara, Alan

Author_Institution

IBM Thomas J. Watson Res. Center, NY

Volume

26

Issue

5

fYear

2006

Firstpage

67

Lastpage

81

Abstract

Optimizing future supercomputing applications will depend on delivering the best performance for a given power budget. To determine the effect on efficiency of application-scaling parameters, this article analyzes system power and performance measurement results for real-world applications exploiting thread- and data-level parallelism on the Blue Gene/L system

Keywords

multi-threading; parallel architectures; parallel machines; performance evaluation; Blue Gene/L system; data-level parallelism; performance optimization; power optimization; supercomputing applications; thread-level parallelism; workload parallelism; CMOS technology; Concurrent computing; Costs; Frequency; Logic devices; Optimized production technology; Parallel processing; Power measurement; Prefetching; System-on-a-chip; Blue Gene/L system; application studies resulting in better multiple-processor systems; architecture; computer system implementation; computer systems organization; interprocessor communications; parallelism; power optimization; processor architectures; super (very large) computers;

fLanguage

English

Journal_Title

Micro, IEEE

Publisher

ieee

ISSN

0272-1732

Type

jour

DOI

10.1109/MM.2006.89

Filename

1709824