Message: Re: How is parallelization implemented using MPI and Geant4? Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Sad Re: How is parallelization implemented using MPI and Geant4? 

Forum: Documentation and Examples
Re: Question How is parallelization implemented using MPI and Geant4? (Krisztian Balla)
Date: 31 Jan, 2009
From: Krisztian Balla <Krisztian Balla>

As expected this does not work. The entire program (including the sensitive detector and the custom run action) is executed for every thread which results in the output file (histogram) being overwritten (lost update).

Looks like one would have to create histogram files for each execution and then cumulate the data on their own.

The other solution would be to somehow send the histograms of all threads (slaves) to the master thread. I can't find functionality for this in the undocumented mpi_interface though.

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 Idea: Re: How is parallelization implemented using MPI and Geant4?   (Krisztian Balla - 31 Jan, 2009)
(_ None: Re: How is parallelization implemented using MPI and Geant4?   (Tom Roberts - 31 Jan, 2009)
(_ Agree: Re: How is parallelization implemented using MPI and Geant4?   (Krisztian Balla - 31 Jan, 2009)
1 None: Re: How is parallelization implemented using MPI and Geant4?   (Tom Roberts - 31 Jan, 2009)
2 Question: Re: How is parallelization implemented using MPI and Geant4?   (Geng - 01 Jul, 2012)
1 None: Re: How is parallelization implemented using MPI and Geant4?   (Youming Yang - 02 Jul, 2012)
... 1 Message(s)
 Add Message Add Message
to: "Re: How is parallelization implemented using MPI and Geant4?"

 Subscribe Subscribe

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


[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]