Premium
A Grid‐based algorithm for the solution of non‐Markovian stochastic Petri nets
Author(s) -
Distefano S.,
Puliafito A.,
Scarpa M.
Publication year - 2007
Publication title -
concurrency and computation: practice and experience
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.309
H-Index - 67
eISSN - 1532-0634
pISSN - 1532-0626
DOI - 10.1002/cpe.1105
Subject(s) - computer science , porting , stochastic petri net , petri net , grid , discretization , distributed computing , parallel computing , markov process , flexibility (engineering) , fault tolerance , algorithm , state space , operating system , mathematics , mathematical analysis , statistics , geometry , software
WebSPN is a modeling tool for the analysis of non‐Markovian stochastic Petri nets that we developed some years ago. Its solution algorithm is based on a discretization of time and an approximation of non‐exponentially distributed firing time transitions by means of the phase‐type distributions. In order to solve the problems related to the management of the state space (which can become very large) we parallelized the solution algorithm through the MPICH libraries. To improve the MPICH parallel implementation with an adequate security management, a more efficient load distribution and fault‐tolerance capabilities, in this paper we propose the porting of WebSPN from the MPI to the Grid computational paradigm. In addition to a better flexibility in accessing computational and storage resources, one of the main advantages is the introduction of a fault recovery system to detect and recover from potential machine faults. The resulting new tool is named GridSPN . Copyright © 2006 John Wiley & Sons, Ltd.