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.

No abstract

No abstract

No abstract

Abstract:These options often have widely varying performance charToday's massively parallel machines are typically message-passing systems consisting of hundreds or thousands of processors. Implementing parallel applications eficiently in this environment is a challenging task, and poor parallel design decisions can be expensive to correct. Tools and techniaues that allow the fast and accurate evaluation acteristics that are functions of numerous system 'and program parameters, and it can be difficult to predict a priori which options are best. Accurate prediction of the performance trade-offs of alternative strategies and of how the performance will change as program parameters change would greatly benefit program developers.of different parallelization strategies would sign@cantly improve the productivity of application developers and increase throughput on parallel architectures. This paper investigates one of the major issues in building tools to compare parallelization strategies: determiningSeveral approaches for the modeling of parallel systems have been Presented that use Markov models Or petri nets [SI, [ll], [12]. Unfortunately, it is difficult to apply these approaches to massively parallel systems or complex paralle1 applications: