Performability analysis of parallel and distributed algorithms

A generalized stochastic Petri net (GSPN) performability model of a parallel and distributed computation is developed. The performance-related activities such as computations and communications are orders of magnitude faster than the component failure and repair activities. Based on the notion of time-scale decomposition, a hierarchy of two levels is defined. At the lower level the performance submodel describes the activities in the application program, while at the higher level the component failure and repair submodel for the underlying architecture defines the current configuration of processors and communication links available for the computation. These two submodels define the reward model needed for performability analysis. Two parallel FFT (fast Fourier transform) algorithms on a hypercube architecture are presented to illustrate the above modeling technique. A general and extended reliability model of the hypercube is also developed. Various performability measures are presented to demonstrate the importance of performability evaluation for mission-critical parallel applications.<<ETX>>

[1]  Michael K. Molloy.  Performance Analysis Using Stochastic Petri Nets , 1982, IEEE Transactions on Computers.

[2]  Stéphane Natkin,et al.  Timed and Stochastic Petri Nets: From the Validation to the Performance of Synchronization Schemes , 1985, Petri Nets and Performance Models.

[3]  C. A. Petri.  Communication with automata , 1966 .

[4]  Ushio Sumita,et al.  Evaluation of minimum and maximum of a correlated pair of random variables via the bivariate laguerre transform , 1986 .

[5]  James Lyle Peterson,et al.  Petri net theory and the modeling of systems , 1981 .

[6]  Marco Ajmone Marsan,et al.  A class of generalized stochastic Petri nets for the performance evaluation of multiprocessor systems , 1984, TOCS.

[7]  Ragnar Huslende,et al.  A combined evaluation of performance and reliability for degradable systems , 1981, SIGMETRICS '81.

[8]  John F. Meyer,et al.  A Performability Solution Method for Degradable Nonrepairable Systems , 1984, IEEE Transactions on Computers.

[9]  John F. Meyer,et al.  Closed-Form Solutions of Performability , 1982, IEEE Transactions on Computers.

[10]  Kang G. Shin,et al.  Modeling of Concurrent Task Execution in a Distributed System for Real-Time Control , 1987, IEEE Transactions on Computers.

[11]  David M. Nicol,et al.  Problem Size, Parallel Architecture, and Optimal Speedup , 1987, J. Parallel Distributed Comput..

[12]  Lorenzo Donatiello,et al.  Analysis of a composite performance reliability measure for fault-tolerant systems , 1987, JACM.

[13]  Kishor S. Trivedi,et al.  NUMERICAL EVALUATION OF PERFORMABILITY AND JOB COMPLETION TIME IN REPAIRABLE FAULT-TOLERANT SYSTEMS. , 1990 .

[14]  Hany H. Ammar,et al.  Performance analysis of degradable multiprocessor system using time scale decomposition of GSPNs , 1988, [1988] Proceedings. The 8th International Conference on Distributed.

[15]  Hany H. Ammar,et al.  Analysis of the Generalized Stochastic Petri Nets by State Aggregation , 1985, PNPM.

[16]  William T. Weeks,et al.  Numerical Inversion of Laplace Transforms Using Laguerre Functions , 1966, JACM.

[17]  Anselmo Lastra,et al.  POET: A Tool for the Analysis of the Performance of Parallel Algorithms , 1988, ICPP.

[18]  L. T. Wu,et al.  Operational models for the evaluation of degradable computing systems , 1982, SIGMETRICS '82.

[19]  Hany H. Ammar,et al.  Time Scale Decomposition of a Class of Generalized Stochastic Petri Net Models , 1989, IEEE Trans. Software Eng..

[20]  Hany H. Ammar,et al.  Hierarchical models for systems reliability, maintainability, and availability , 1987 .

[21]  Alan Norton,et al.  Parallelization and Performance Analysis of the Cooley–Tukey FFT Algorithm for Shared-Memory Architectures , 1987, IEEE Transactions on Computers.

[22]  Michael K. Molloy,et al.  On the integration of delay and throughput measures in distributed processing models , 1981 .

[23]  Asser N. Tantawi,et al.  Evaluation of Performability for Degradable Computer Systems , 1987, IEEE Transactions on Computers.

[24]  Philip Heidelberger,et al.  Analysis of Performability for Stochastic Models of Fault-Tolerant Systems , 1986, IEEE Transactions on Computers.

[25]  Ajmone MarsanMarco,et al.  A class of generalized stochastic Petri nets for the performance evaluation of multiprocessor systems , 1984 .

[26]  Krishnan Padmanabhan,et al.  Reliability of the Hypercube , 1988, ICPP.

[27]  Kishor S. Trivedi,et al.  Performability Analysis: Measures, an Algorithm, and a Case Study , 1988, IEEE Trans. Computers.

[28]  John F. Meyer,et al.  On Evaluating the Performability of Degradable Computing Systems , 1980, IEEE Transactions on Computers.