Re: about checkpoint support

From: Paul H. Hargrove (PHHargrove_at_lbl_dot_gov)
Date: Thu Jan 08 2009 - 11:32:02 PST

  • Next message: Paul H. Hargrove: "Announcing the release of BLCR 0.8.0"
    jette1@llnl.gov wrote:
    >> Dear Morris,
    >>
    >> I am working on integrating BLCR with SLURM. The progress is delayed 
    >> because I am busy this term. I write another checkpoint plugin, 
    >> checkpoint/blcr. Also the checkpoint plugin API was changed. I am 
    >> testing and debugging the code. I wonder whether this may be accepted 
    >> by SLURM. If OK, I can write an detailed document of the work after I 
    >> complete it.
    >>
    >> Best Regards,
    >> Hongjia Cao
    >
    >
    > Dear Hongjia,
    >
    > The BLCR developers have already performed some integration of
    > BLCR with SLURM for SiCortex (a US computer vendor). I am including
    > the people who are involved in the work in this response so that
    > we can see what capabilities exist today and what capabilities
    > would be valuable to add in the future. If anything can be done
    > to SLURM to improve its integration with BLCR, that could certainly
    > take the form of a new SLURM plugin. It would be great if you
    > could do that work.
    >
    
    Hongjia,
     
    I am afraid that Moe (Morris) either misunderstood our discussions in 
    November, or is misremembering them.
    
    Our BLCR work has *not* yet done any integration with SLURM.  I 
    mentioned to Moe that we had been doing BLCR integration work with TORQUE.
    I also described to Moe work we had done to integrate SLURM on the 
    SiCortex platform with our Berkeley UPC compiler.
    So, with respect to BLCR+SLURM integration there are not any 
    "capabilities that exist today" from which to expand.
    
    I do share Moe's interest in you completing work on a SLURM plugin for 
    BLCR integration.  If there are any questions about BLCR that I could 
    answer for you, feel free to email our list: checkpoint_at_lbl_dot_gov.
    
    -Paul
    
    -- 
    Paul H. Hargrove                          PHHargrove_at_lbl_dot_gov
    Future Technologies Group                 Tel: +1-510-495-2352
    HPC Research Department                   Fax: +1-510-486-6900
    Lawrence Berkeley National Laboratory     
    

  • Next message: Paul H. Hargrove: "Announcing the release of BLCR 0.8.0"