NAS Parallel Benchmarks
The NAS Parallel Benchmarks (NPB) are a set of benchmarks targetting performance evaluation of highly parallel supercomputers. They are developed and maintained by the NASA Advanced Supercomputing Division (NAS) (formerly the NASA Numerical Aerodynamic Simulation Program) based at the NASA Ames Research Center. NAS solicits performance results for NPB from all sources.[1]
History
Motivation
Traditional benckmarks that existed before NPB, such as the Livermore loops, the LINPACK Benchmark and the NAS Kernel Benchmark Program, were usually specialized for vector computers. They generallly suffered from inadequacies including parallelism-impeding tuning restrictions and insufficient problem sizes, which rendered them inappropriate for highly parallel systems. Equally unsuitable were full-scale application benchmarks due to high porting cost and unavailability of automatic software parallelization tools.[2] As a result, NPB were released in 1991 to address the ensuing lack of benchmarks applicable to highly parallel machines.
NPB 1
The first specification of NPB recognized that the benchmarks should feature
- new parallel-aware algorithmic and software methods,
- genericness and architecture neutrality,
- easy verifiability of correctness of results and performance figures,
- capability of accomodating new systems with increased power,
- and ready distributability.
In the light of these guidelines, it was deemed the only viable approach to use a collection of "paper and pencil" benchmarks that specified a set of problems only algorithmically and left most implementation details to the implementor's discretion under certain necessary limits. Sample codes written in Fortran 77 were supplied but not intended for benchmarking purposes.[2]
NPB 2
Since its release, NPB 1 displayed two major weaknesses. Firstly, due to its "paper and pencil" style of specification, computer vendors usually highly tuned their implementations so that their performance became difficult for scienctific programmers to attain. Moreover, many of these implementation were proprietary and not publicly available, effectively concealing their optimizing techniques. Secondly, problem sizes of NPB 1 lagged behind the development of supercomputers as the latter continued to evolve.[3]
References
- ^ "NAS Parallel Benchmarks Changes". NASA Advanced Supercomputing Division. Retrieved 2009-02-23.
- ^ a b Baily, D.; Barscz, E.; Barton, J.; Browning, D.; Carter, R.; Dagum, L.; Fatoohi, R.; Fineberg, S.; Frederickson, P.; Lasinski, T.; Schreiber, R.; Simon, H.; Venkatakrishnan, V.; Weeratunga, S. (March 1994), "The NAS Parallel Benchmarks" (PDF), NAS Technical Report RNR-94-007
{{citation}}: CS1 maint: year (link) - ^ Bailey, D.; Harris, T.; Saphir, W.; van der Wijngaart, R.; Woo, A.; Yarrow, M. (December 1995), "The NAS Parallel Benchmarks 2.0" (PDF), NAS Technical Report NAS-95-020
{{citation}}: CS1 maint: year (link)
External links
- NAS Parallel Benchmarks Changes (official website)
- NAS Kernel Benchmark Program
 | This software article is a stub. You can help Wikipedia by expanding it. |