DocumentCode :
3204898
Title :
Architectural Constraints to Attain 1 Exaflop/s for Three Scientific Application Classes
Author :
Bhatele, Abhinav ; Jetley, Pritish ; Gahvari, Hormozd ; Wesolowski, Lukasz ; Gropp, William D. ; Kale, Laxmikant
Author_Institution :
Dept. of Comput. Sci., Univ. of Illinois at Urbana-Champaign, Urbana, IL, USA
fYear :
2011
fDate :
16-20 May 2011
Firstpage :
80
Lastpage :
91
Abstract :
The first Teraflop/s computer, the ASCI Red, became operational in 1997, and it took more than 11 years for a Petaflop/s performance machine, the IBM Roadrunner, to appear on the Top500 list. Efforts have begun to study the hardware and software challenges for building an exascale machine. It is important to understand and meet these challenges in order to attain Exaflop/s performance. This paper presents a feasibility study of three important application classes to formulate the constraints that these classes will impose on the machine architecture for achieving a sustained performance of 1 Exaflop/s. The application classes being considered in this paper are -- classical molecular dynamics, cosmological simulations and unstructured grid computations (finite element solvers). We analyze the problem sizes required for representative algorithms in each class to achieve 1 Exaflop/s and the hardware requirements in terms of the network and memory. Based on the analysis for achieving an Exaflop/s, we also discuss the performance of these algorithms for much smaller problem sizes.
Keywords :
astronomy computing; biology computing; cosmology; finite element analysis; grid computing; mathematics computing; molecular biophysics; multiprocessing systems; ASCI Red; IBM Roadrunner; architectural constraints; cosmological simulations; exaflops performance; exascale machine; finite element solvers; machine architecture; molecular dynamics; petaflops performance machine; representative algorithm; scientific application classes; teraflops computer; unstructured grid computations; Atomic measurements; Bandwidth; Biological system modeling; Computational modeling; Equations; Mathematical model; Memory management;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Parallel & Distributed Processing Symposium (IPDPS), 2011 IEEE International
Conference_Location :
Anchorage, AK
ISSN :
1530-2075
Print_ISBN :
978-1-61284-372-8
Electronic_ISBN :
1530-2075
Type :
conf
DOI :
10.1109/IPDPS.2011.18
Filename :
6012827
Link To Document :
بازگشت