Predicting Runtime Performance Bounds of Expanded Parallel Discrete Event Simulations

Abstract

Abstract-Predicting and analyzing runtime performance characteristics is a vital step in the development process of parallel discrete event simulations. For instance, model developers need to identify and eliminate performance bottlenecks within a simulation model in order to derive a model structure that aids parallel execution. Similarly, developers of parallel simulation frameworks require means of assessing the efficiency of the framework. In this paper, we present a performance prediction methodology that computes the best possible performance bound for expanded parallel discrete event simulations in the context of our Horizon simulation framework. The methodology builds upon a linear program which calculates an optimal event execution schedule for a given simulation and a set of CPUs. In order to mitigate the complexity of this NP-complete scheduling problem, we introduce performance optimizations and relaxations of the linear program.

Evaluation and Improvement of Parallel Discrete Event Simulation Performance Predictions: A Rough-Set-based Approach

“…The method described in [7], uses a hybrid approach to define the theoretical limit of the execution improvement for conservative synchronization protocol. In the trace-based part, the method analyses only a definite part of the sequence of events of the simulation model; while in an analytical modeling part, for the calculation of the lower bound on the runtime of the parallel simulation, a simplified scheduling problem model and the linear programming approach of is applied.…”

“…Development of simulation models having high runtime performance features in a PDES execution environment is not an easy task even today, mainly because it is hard to tell the parallel execution features of a model, for example, possible resource capacity bottlenecks, in the development phase [7]. The simulation performance prediction of PDES execution can help to realize higher performance by providing preliminary knowledge about the behavior of the simulation model in the parallel execution environment [7,8,9,10,11].…”

“…Development of simulation models having high runtime performance features in a PDES execution environment is not an easy task even today, mainly because it is hard to tell the parallel execution features of a model, for example, possible resource capacity bottlenecks, in the development phase [7]. The simulation performance prediction of PDES execution can help to realize higher performance by providing preliminary knowledge about the behavior of the simulation model in the parallel execution environment [7,8,9,10,11]. The PDES performance prediction methods can support the performance increase throughout the whole modeling and simulation process including model development support, simulation setup and evaluation phases and taking into account different simulation framework specifics too [7].…”

“…The simulation performance prediction of PDES execution can help to realize higher performance by providing preliminary knowledge about the behavior of the simulation model in the parallel execution environment [7,8,9,10,11]. The PDES performance prediction methods can support the performance increase throughout the whole modeling and simulation process including model development support, simulation setup and evaluation phases and taking into account different simulation framework specifics too [7].…”

On using Markov Decision Processes to Model Integration Solutions for Disparate Resources in Software Ecosystems

“…With the model proposed in this study, it was possible to understand and improve the performance of the production system; furthermore, it was found that the discrete-event simulation is capable of analysing the behaviour of complex and sophisticated systems. Kunz et al (2011) present a performance prediction methodology that calculates the best possible performance bound for expanded parallel discreteevent simulations. According to the authors, predicting and analysing runtime performance features and understanding its behaviour is an important step in the development process of parallel discrete-event simulations.…”