[OMPI users] Checkpoint and restart

2008-07-22 Thread Gabriele Fatigati
Dear OpenMPI developers,
i have some questions about application robustness. , I have read that in
OpenMPI development trunk version will have implemented techniques of
checkpoint and restart. But, i need to know ho this techniques works. Are
they internal functions to insert in my code, external tool or other?

And, when i restart my application, is it possible to modify the initial
number of processors?

-- 
Gabriele Fatigati

CINECA Systems & Tecnologies Department

Supercomputing Group

Via Magnanelli 6/3, Casalecchio di Reno (BO) Italy

www.cineca.it Tel: +39 051 6171722

g.fatig...@cineca.it


Re: [OMPI users] Checkpoint and restart

2008-07-22 Thread Josh Hursey


On Jul 22, 2008, at 5:43 AM, Gabriele Fatigati wrote:


Dear OpenMPI developers,
i have some questions about application robustness. , I have read  
that in OpenMPI development trunk version will have implemented  
techniques of checkpoint and restart. But, i need to know ho this  
techniques works. Are they internal functions to insert in my code,  
external tool or other?


The checkpoint/restart technique is transparent so there are no  
functions to insert into your code. The functionality is described in  
the user document on the wiki:

  https://svn.open-mpi.org/trac/ompi/wiki/ProcessFT_CR




And, when i restart my application, is it possible to modify the  
initial number of processors?


Not at this time. You must restart with the same number of processes  
(-np) as you initially started with.


-- Josh




--
Gabriele Fatigati

CINECA Systems & Tecnologies Department

Supercomputing Group

Via Magnanelli 6/3, Casalecchio di Reno (BO) Italy

www.cineca.it Tel: +39 051 6171722

g.fatig...@cineca.it
___
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users