NAS Parallel Benchmarks: Difference between revisions

Revision as of 00:49, 22 February 2009

Revision as of 03:20, 24 February 2009

Line 1:

The '''NAS Parallel Benchmarks''' ('''NPB''') are a set of [[benchmark (computing)|benchmark]]s targetting performance evaluation of highly parallel [[supercomputer]]s. They are developed and maintained by the [[NASA Advanced Supercomputing facility|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.<ref name=npbweb>{{cite web

|title=NAS Parallel Benchmarks Changes

|url=http://www.nas.nasa.gov/Resources/Software/npb.html

|publisher=[[NASA Advanced Supercomputing facility|NASA Advanced Supercomputing Division]]

|accessdate=2009-02-23

}}</ref>

==DevelopmentHistory==

===Motivation===

Traditional benckmarks that existed before the NPB, such as the [[Livermore loops]], the [[LINPACK|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.<ref name=rnr94007>D. Bailey, E. Barscz, J. Barton, D. Browning, R. Carter, L. Dagum, R. Fatoohi, S. Fineberg, P. Frederickson, T. Lasinski, R. Schreiber, H. Simon, V. Venkatakrishnan, S. Weeratunga, [http://www.nersc.gov/~simon/Papers/NASA/RNR-94-007.pdf The NAS Parallel Benchmarks], NAS Technical Report RNR-94-007, NASA Ames Research Center, Moffett Field, CA, 1994.</ref> As a result, the NPB were created in the early 1990s to address the ensuing lack of benchmarks applicable to highly parallel machines.{{Citation

|last1=Baily|first1=D.

|last2=Barscz|first2=E.

|last3=Barton|first3=J.

|last4=Browning|first4=D.

|last5=Carter|first5=R.

|last6=Dagum|first6=L.

|last7=Fatoohi|first7=R.

|last8=Fineberg|first8=S.

|last9=Frederickson|first9=P.

|last10=Lasinski|first10=T.

|last11=Schreiber|first11=R.

|last12=Simon|first12=H.

|last13=Venkatakrishnan|first13=V.

|last14=Weeratunga|first14=S.

|contribution=The NAS Parallel Benchmarks

|contribution-url=http://www.nersc.gov/~simon/Papers/NASA/RNR-94-007.pdf

|title=NAS Technical Report RNR-94-007

|year=March 1994

}}</ref> As a result, NPB were released in 1991 to address the ensuing lack of benchmarks applicable to highly parallel machines.

===GuidelinesNPB 1===

The first specification of the NPB recognizesrecognized that the followingbenchmarks developmentshould guidelines:feature

* Use of new parallel-aware algorithmic and software methods,

* Genericnessgenericness and architecture neutrality,

* Easyeasy verifiability of correctness of results and performance figures,

* Capabilitycapability of accomodating new systems with increased power,

* Readyand ready distributability.

In the light of these guidelines, it was deemed the only viable approach to use a setcollection of "paper and pencil" benchmarks, i.e.,that specifyspecified a set of problems only algorithmically and leaveleft most implementation details to the implementor's discretion under certain less restrictivenecessary limits. Sample codes written in [[Fortran 77]] were supplied but not intended for benchmarking purposes.<ref name=rnr94007/>

===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.<ref name=nas95020>{{Citation

|last1=Bailey|first1=D.

|last2=Harris|first2=T.

|last3=Saphir|first3=W.

|last4=van der Wijngaart|first4=R.

|last5=Woo|first5=A.

|last6=Yarrow|first6=M.

|contribution=The NAS Parallel Benchmarks 2.0

|contribution-url=http://www.nas.nasa.gov/News/Techreports/1995/PDF/nas-95-020.pdf

|title=NAS Technical Report NAS-95-020

|year=December 1995

}}

</ref>

==References==

Line 19 ⟶ 58:

== External links ==

* [http://www.nas.nasa.gov/Software/NPB/ NAS Parallel Benchmarks Changes] (official website)

* [http://www.netlib.org/benchmark/nas NAS Kernel Benchmark Program]