Message: Re: Clarifying the correct way to run MT jobs with MPI on a cluster with multiple nodes Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Feedback Re: Clarifying the correct way to run MT jobs with MPI on a cluster with multiple nodes 

Forum: Multithreading
Re: Question Clarifying the correct way to run MT jobs with MPI on a cluster with multiple nodes (Sam Blake)
Re: Question Re: Clarifying the correct way to run MT jobs with MPI on a cluster with multiple nodes (Svetlana Shasharina)
Date: 07 Apr, 2016
From: Andrea Dotti <Andrea Dotti>

Hello, if you have compiled Geant4 with multi-threading support and you have migrated your application to use multi-threading (see: https://twiki.cern.ch/twiki/bin/view/Geant4/QuickMigrationGuideForGeant4V10) you can use multiple cores on your machine to speedup the simulation. You specify the number of threads via the ui command: /run/numberOfThreads <N> or, to use the maximum available cores: /run/useMaximumLogicalCores Use these commands BEFORE /run/initialize

If you want to scale across multiple nodes, you need to use MPI. There are examples under examples/extended/parallel/MPI. This however is a bit more advanced.

A final note: the number of events is an integer, it cannot be large than ~2*10^9 (see https://en.wikibooks.org/wiki/C_Programming/C_Reference/limits.h). If you need more events you need to do multiple runs.

Andrea

 Add Message Add Message
to: "Re: Clarifying the correct way to run MT jobs with MPI on a cluster with multiple nodes"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews