Abstract
HPL is a parallel Linpack benchmark package widely adopted in massive cluster system performance test. On HPL data layout among processors, a law to determine block size NB theoretically, which breaks through dependence on trial-and-error experiments, is found based on in-depth analysis of blocked parallel solution algorithm of linear algebra equations and implementation mechanics in HPL. According to that law, an emulation model to toughly estimate HPL execution time is constructed. Verified by real system, the model is used to do some scientific prevision on the benefits to Linpack test brought by intending system improvement, such as respectively memory size increase, communication bandwidth increase and so on. It is expected to conduce to direct system improvement on optimizing HPL test in the future.
This research is supported by Chinese National High-tech Research and Development (863) Program (grants 2003AA 1Z2070) and by the foundation of Knowledge Innovation Program (grants 20036040), Chinese Academy of Sciences (CAS).
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Dongarra, J.J., Luszczek, P., Petitet, A.: The LINPACK Benchmark: Past, Present, and Future, Concurrency and Computation: Practice and Experience 15 (2003)
Petitet, A., Whaley, R.C., Dongarra, J., Cleary, A.: HPL – A Portable Implementation of the High-Performance Linpack Benchmark for Distributed-Memory Computers, http://www.netlib.org/benchmark/hpl/
Meuer, H.W., Strohmaier, E., Dongarra, J.J., Simon, H.D.: Top500 Supercomputer Sites, 17th edn., November 2 (2001), The report can be downloaded from, http://www.netlib.org/benchmark/top500.html
Zhang, B.L., et al.: Theory and method of numeric parallel computing. National defense industry press, Beijing (1999)
Lin, C.S.: Numeric computing method (Column A). Science press, Beijing (1998)
Chen, G.L.: Parallel computing: structure, algorithm, programming (modified version), p. 8. Advanced education press, Beijing (2003)
Sun, Z.Z.: Numeric analysis, 2nd edn., p. 1. South-east university press, Nanjing (2002)
http://www.cs.utk.edu/~dongarra/WEB-PAGES/SPRING-2000/lect08.pdf
Wenli, Z., Jianping, F., Mingyu, C.: Efficient Determination of Block Size NB for Parallel Linpack Test. In: Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems (PDCS 2004). MIT, Cambridge (2004) (received)
Caron, E., Utard, G.: On the performance of parallel factorization of out-of-core matrices. Parallel computing 30, 357–375 (2004)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhang, W., Chen, M., Fan, J. (2004). HPL Performance Prevision to Intending System Improvement. In: Cao, J., Yang, L.T., Guo, M., Lau, F. (eds) Parallel and Distributed Processing and Applications. ISPA 2004. Lecture Notes in Computer Science, vol 3358. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30566-8_90
Download citation
DOI: https://doi.org/10.1007/978-3-540-30566-8_90
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-24128-7
Online ISBN: 978-3-540-30566-8
eBook Packages: Computer ScienceComputer Science (R0)