DocumentCode
669930
Title
Quantifying the energy cost of data movement in scientific applications
Author
Kestor, Gokcen ; Gioiosa, Roberto ; Kerbyson, Darren J. ; Hoisie, Adolfy
Author_Institution
Pacific Northwest Nat. Lab., Richland, WA, USA
fYear
2013
fDate
22-24 Sept. 2013
Firstpage
56
Lastpage
65
Abstract
In the exascale era, the energy cost of moving data across the memory hierarchy is expected to be two orders of magnitude higher than the cost of performing a double-precision floating point operation. Despite its importance, the energy cost of data movement in scientific applications has not be quantitatively evaluated even for current systems.
Keywords
data handling; floating point arithmetic; natural sciences computing; storage management; DoE exascale codesign centers; data movement; data prefetcher; double-precision floating point operation; energy cost quantification; exascale era; external power measurements; highly-tuned micro-benchmarks; internal power measurements; processor registers; science office; scientific applications; stalled cycles; state-of-the-art compute node systems; Benchmark testing; Energy consumption; Energy measurement; Fans; Power demand; Prefetching; Registers;
fLanguage
English
Publisher
ieee
Conference_Titel
Workload Characterization (IISWC), 2013 IEEE International Symposium on
Conference_Location
Portland, OR
Print_ISBN
978-1-4799-0553-9
Type
conf
DOI
10.1109/IISWC.2013.6704670
Filename
6704670
Link To Document