Performance of parallel computers for spectral atmospheric models

Foster, I T; Toonen, Brian; Worley, P.H.

doi:10.2172/113769

Cited by 9 publications

(14 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Efficient parallelizations of PSTSWM exploit two-dimensional decompositions of the domain, parallelizing both the FFTs and LTs [4]. The studies described in Sect.…”

Section: Full Simulation Performancementioning

confidence: 99%

“…Communication Cray Research T3D [4], [14]. Although these machines are all c'classic" distributedmemory MIMD parallel systems, their performance characteristics differ significantly, as do their optimal tuning techniques and parameters.…”

Section: Introductionmentioning

confidence: 99%

“…PSTSWM was developed to evaluate strategies for parallelizing spectral global atmospheric circulation models [4], [5], and has imbedded a large number of parallel algorithm options. Among these options are numerous choices for the communication protocols used to implement the different parallel algorithms and numerous choices of message-passing layer.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Impact of Communication Protocol on Performance

Worley¹

1999

Self Cite

View full text Add to dashboard Cite

“…Efficient parallelizations of PSTSWM exploit two-dimensional decompositions of the domain, parallelizing both the FFTs and LTs [4]. The studies described in Sect.…”

Section: Full Simulation Performancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Impact of Communication Protocol on Performance

Worley¹

1999

Self Cite

View full text Add to dashboard Cite

“…Multiple parallel algorithms and algorithm variants are embedded in PSTSWM, allowing good algorithms to be identified from empirical studies. The results of the studies using PSTSWM have been extremely useful [12,13]; however, PSTSWM took over two years to develop and the experiments required to identify the best algorithms are time-consuming. We hope that performance models would be simpler to adapt to proposed changes in the application codes and could be used to quickly examine the effect of running on new machines or with different problem or machine parameters.…”

Section: Introductionmentioning

confidence: 99%

Performance modeling for SPMD message-passing programs

Brehm

Worley

Madhukar

1998

Concurrency: Pract. Exper.

View full text Add to dashboard Cite

Today's massively parallel machines are typically message-passing systems consisting of hundreds or thousands of processors. Implementing parallel applications efficiently in this environment is a challenging task, and poor parallel design decisions can be expensive to correct. Tools and techniques that allow the fast and accurate evaluation of different parallelization strategies would significantly improve the productivity of application developers and increase throughput on pardel architectures. This paper investigates one of the major issues in building tools to compare parallelization strategies: determining what type of performance models of the application code and of the computer system are sufficient for a fast and accurate comparison of different strategies. The paper is built around a case study employing the Performance Prediction Tool (PerPreT) to predict performance of the Parallel Spectral Transform Shallow Water Model code (PSTSWM) on the Intel Paragon.PSTSWM is a parallel application code that was designed to evaluate different pard e l strategies for the spectral transform method as it is used in climate modeling and weather forecasting. Multiple parallel algorithms and algorithm variants are embedded in the code. PerPreT uses a relatively simple algebraic model to predict execution time for SPMD (Single Program Multiple Data) parallel applications. Applications are modeled through parameterized formulae for communication and computation, where the parameters include the problem size, the number of processors used to execute the program, and system characteristics (e+, setup times for communication, link bandwidth, and sustained computing performance per processor).In this paper we describe performance models that predict the performance of the different algorithms in PSTSWM accurately enough to allow them to be compared, establishing the feasibility of such a demanding application of performance modeling. We also discuss issues in generating and validating the performance models, emphasizing the practical importance of tools such as PerPreT in such studies.

show abstract

“…Algorithm comparison results are described in [6]. Benchmark results are described in [18] and [5]. The benchmarking philosophy inspired by PSTSWM is described in [17].…”

Section: Introductionmentioning

confidence: 99%