Subject: [Condor-users] Fault tolerance MPI and CONDOR
Hi all, I have a doubt about the fault tolerance that manages
to integrate CONDOR with MPI , what happens when you're already running a
job and a machine fails, the manager is able to re-distribute the load
that was running the job presented fault or lost
overall job performance.
I
appreciate the collaboration, you can give me.
-- Omaira Galindo Parra Estudiante de Ingeniería de Sistemas y Computación UPTC Tunja