Implementation of a Monte Carlo code on a parallel computer system
This article describes the implementation of a Monte Carlo based code for a multiple-transputer system (Edinburgh Concurrent Supercomputer). The application is related to measuring the radiation dose absorbed in cylindrical dosimeters. The evolution of particles is supposed to be described by a Markov chain. The conclusion of the article is that there is a linear increase of speed, proportional to the number of transputers involved in the simulation process. Also, the author proposes the generalization of the basic techniques used in the work on other parallel computer systems which have similar operational structures. Three sequential programs are identified: a control process to generate the tasks, a number of simulation processes to consume the tasks and to generate results, an analysis process to collect and analyze results. All these programs have flow diagrams in the article. Special tools are used for programming in transputer environment (CST) and interface subroutines are described.
- Tradeoffs in granularity and parallelization for a Monte Carlo shower simulation code
- An object-oriented implementation of a parallel Monte Carlo code for radiation transport
- scientific article; zbMATH DE number 1163493 (Why is no real title available?)
- scientific article; zbMATH DE number 1506986 (Why is no real title available?)
- scientific article; zbMATH DE number 2097481 (Why is no real title available?)
- Parallelization of a Monte Carlo particle transport simulation code
This page was built for publication: Implementation of a Monte Carlo code on a parallel computer system
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1334552)