Re: OpenMosix and Checkpointing

Daniel-
  This seems to becoming a often recently, with people assuming that
because openMosix does migration it it can do check pointing. Would
you mind enlightening me as to why that is.

 I would like to clear up one possible error in your impression of
openMosix, there is no subset of features for HPC, that is all
openMosix is about. If some wants any kind of HAC deployment openMosxi
would be counter productive as a single node failure can bring make
the whole cluster hiccup, and at the very least toats any jobs that
where touching the failed node at time of failure.

Evan

On 4/4/06, Daniel Marques <[email protected]> wrote:
> Hello.
>
> I'm a computer science researcher who has been studying checkpointing.
>
> I'm very new to OpenMosix (so bare with me here), but am interested in
> understanding how OpenMosix could be integrated with some of my
> group's work in checkpointing MPI applications. (see "Implementation
> and Evaluation of a Scalable Application-level Checkpoint-Recovery
> Scheme for MPI Programs" in Supercomputing 2004, if you are
> interested.)
>
> That work focuses on the coordination of MPI processes to take a
> consistent checkpoint.  We are interested in using the OpenMosix
> software to checkpoint the individual process states.
>
> From reading the OpenMosix documentation, it seems that checkpointing
> is not provided by OpenMosix, but rather with the CHPOX module.
>
> It seems that if OpenMosix supports process migration, it should be
> possible to extend it to support checkpointing.  Is that what CHPOX
> does, or is it an independent approach to taking checkpoints?  If the
> later, how hard would it be to extend OpenMosix itself?
> (We are only concerned with HPC type apps, i.e. a subset of what
> OpenMosix can migrate.)
>
> Like I said, I am just beginning to learn about your software, so I'd
> appreciate any and all information and pointers in the right
> direction.
>
> (Please reply directly or CC me on the emails, as I haven't subscribed
> to this list yet.)
>
> Thanks.
>
> Dan
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmdlnk&kid0944&bid$1720&dat1642
> _______________________________________________
> openMosix-devel mailing list
> openMosix-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/openmosix-devel
>


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid0944&bid$1720&dat1642