Performance prediction for complex parallel applications
作者： J. Brehm , P.H. Worley
DOI: 10.2172/467122
关键词:
摘要: Today's massively parallel machines are typically message-passing systems consisting of hundreds or thousands processors. Implementing applications efficiently in this environment is a challenging task, and poor design decisions can be expensive to correct. Tools techniques that allow the fast accurate evaluation different parallelization strategies would significantly improve productivity application developers increase throughput on architectures. This paper investigates one major issues building tools compare strategies: determining what type performance models code computer system sufficient for comparison strategies. The built around case study employing Performance Prediction Tool (PerPreT) predict Parallel Spectral Transform Shallow Water Model (PSTSWM) Intel Paragon.
core.ac.uk
osti.gov
sci-hub.st 参考文章(10)
Jürgen Brehm, Manish Madhukar, Evgenia Smirni, Larry Dowdy, PerPreT - A Performance Prediction Tool for Massive Parallel Sysytems MMB '95 Proceedings of the 8th International Conference on Modelling Techniques and Tools for Computer Performance Evaluation: Quantitative Evaluation of Computing and Communication Systems. pp. 284- 298 ,(1995) , 10.1007/BFB0024322
H. Wabnig, G. Haring, PAPS: the parallel program performance prediction toolset Proceedings of the 7th international conference on Computer performance evaluation : modelling techniques and tools: modelling techniques and tools. pp. 284- 304 ,(1994) , 10.1007/3-540-58021-2_16
Heidelberger, Trivedi, Analytic Queueing Models for Programs with Internal Concurrency IEEE Transactions on Computers. ,vol. 32, pp. 73- 82 ,(1983) , 10.1109/TC.1983.1676125
Thomasian, Bay, Analytic Queueing Network Models for Parallel Processing of Task Systems IEEE Transactions on Computers. ,vol. 35, pp. 1045- 1054 ,(1986) , 10.1109/TC.1986.1676712
Ian T. Foster, Patrick H. Worley, Parallel Algorithms for the Spectral Transform Method SIAM Journal on Scientific Computing. ,vol. 18, pp. 806- 837 ,(1997) , 10.1137/S1064827594266891
JÜRGEN BREHM, PATRICK H. WORLEY, MANISH MADHUKAR, Performance modeling for SPMD message-passing programs Concurrency and Computation: Practice and Experience. ,vol. 10, pp. 333- 357 ,(1998) , 10.1002/(SICI)1096-9128(19980425)10:5<333::AID-CPE321>3.0.CO;2-X
M. Parashar, S. Hariri, Compile-time performance prediction of HPF/Fortran 90D IEEE Parallel & Distributed Technology: Systems & Applications. ,vol. 4, pp. 57- 73 ,(1996) , 10.1109/88.481665
T. Fahringer, Estimating and optimizing performance for parallel programs IEEE Computer. ,vol. 28, pp. 47- 56 ,(1995) , 10.1109/2.471179
M. Calzarossa, G. Serazzi, Workload characterization: a survey Proceedings of the IEEE. ,vol. 81, pp. 1136- 1150 ,(1993) , 10.1109/5.236191
S.R. Sarukkai, P. Mehra, R.J. Block, Automated scalability analysis of message-passing parallel programs IEEE Parallel & Distributed Technology: Systems & Applications. ,vol. 3, pp. 21- 32 ,(1995) , 10.1109/88.473611