DocumentCode :
2522515
Title :
Sizing equivalent performing large-scale systems using modeling
Author :
Kerbyson, Darren J. ; Hoisie, Adolfy
Author_Institution :
Performance & Archit. Lab., Los Alamos Nat. Lab., NM, USA
fYear :
2003
fDate :
37819
Firstpage :
46
Lastpage :
55
Abstract :
This work analyses the relative performance of the Earth Simulator to systems constructed from HP AlphaServer nodes such as ASCI Q. The Earth Simulator uses vector processing nodes interconnected using a single-stage cross-bar network. The AlphaServer systems use microprocessors with deep memory hierarchies and are typically interconnected using the Quadrics Elan3 fat-tree network. The performance that can be achieved on a system result from an interplay of the characteristics of the system, the requirements of the workload, and scaling behavior of the two. Detailed performance models are used here to predict the performance of two codes representative of ASCI computations, namely SAGE and Sweep3D. The models are used to compare the performance of the two systems and in particular size an AlphaServer system that achieves the same performance as the Earth Simulator on SAGE and Sweep3D.
Keywords :
digital simulation; distributed shared memory systems; microprocessor chips; parallel machines; performance evaluation; ASCI Q; ASCI computations; AlphaServer systems; Earth Simulator; HP AlphaServer nodes; Quadrics Elan3 fat-tree network; SAGE; Sweep3D; large-scale systems; memory hierarchies; microprocessors; modeling; performance evaluation; single-stage cross-bar network; sizing equivalent; vector processing nodes; Analytical models; Computational modeling; Earth; Hardware; Laboratories; Large-scale systems; Microprocessors; Performance analysis; Predictive models; Procurement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Innovative Architecture for Future Generation High-Performance Processors and Systems, 2003
ISSN :
1537-3223
Print_ISBN :
0-7695-2019-7
Type :
conf
DOI :
10.1109/IWIA.2003.1262782
Filename :
1262782
Link To Document :
بازگشت