Re: RootMpi

Hello Tim.
this is a great question.
The idea is to write a performance test that measures the entire data transmission, double precision, arrays and objects and regarding the transmission of objects we need to get the time in serializing and deserializing.
This should be for peer to peer and collective communication also Mpi's reduce operations and for synchronous and asynchronous communication. Let me write something soon to measure the performance of the integration's algorithms, the data will be put in the blog and I will tell you this in the mailing list.

Thanks Tim.
best wishes

At the moment I don't have a performance test, but let me to write something

2012/3/29 Tim dos Santos <dos.santos_at_physik.uni-wuppertal.de>

> Hi Omar,
>
> I'm curious, have you recorded some performance measurements yet, showing
> the net benefit of MPI'ing e.g. an integration to n Cores, and the like?
>
> Keep up the good work!
>
> Tim
>
>
>
> Am 28.03.2012 22:18, schrieb Omar Andrés Zapata Mesa:
>
> Hello Rooters
> The GSoC proposal was submitted, the information in the blog was updated
> http://rootmpi.blogspot.com/
> You can download now the debian package to test the RootMpi's prototype
> http://rootmpi.blogspot.com/2012/03/installing-root-libraries.html
> Some feedback should be nice.
> Thanks!
>
> --
> Omar Andres Zapata Mesa
> Fundamental Interaction Phenomenology Gruop (Gfif)
> Head Developer
> http://gfif.udea.edu.co
> Gfif Developers (Gfif Dev)
> Division of computer science
> System Engineering Student
> Universidad de Antioquia At Medellin - Colombia
> Usuario Linux #490962
>
>
Received on Thu Mar 29 2012 - 20:44:37 CEST