A PARALLELISM ANALYZER FOR CONSERVATIVE PARALLEL SIMULATION

Most small-scale simulation applications are implemented by sequential simulation techniques, As the problem size increases, however, sequen tial techniques may be unable to manage the time complexity of the sim ulation applications adequately, It is natural to consider re-implemen ting the corresponding large-scale simulations using parallel techniqu es, which have been reported to be successful in reducing the time com plexity for several examples, However, parallel simulation may not be effective for every application, Since the implementation of parallel simulation for an application is usually very expensive, it is require d to investigate the performance of parallel simulation for a particul ar application before re-implementing the simulation, The Chandy-Misra parallel, discrete-event simulation paradigm has been utilized in man y large-scale simulation experiments, and several significant extensio ns have been based on it, Hence the Chandy-Misra protocol is adopted h ere as a basic model of parallel simulation to which our performance p rediction techniques are applied, For an existing sequential simulatio n program based on the process interaction model, this paper proposes a technique for evaluating Chandy-Misra parallel simulation without ac tually implementing the parallel program, The idea is to insert parall elism analysis code into the sequential simulation program, When the m odified sequential program is executed, the time complexity of the par allel simulation based on the Chandy-Misra protocol is computed, Our t echnique has been used to determine whether a giant Signaling System 7 simulation (sequential implementation) should be re-implemented using the parallel simulation approach.