Re: [Condor-users] Checkpointing in vanilla

[Matt Hope <matthew.hope@xxxxxxxxx>]

On 1/19/06, Thomas Materna <materna@xxxxxxxxxxxxx> wrote:
>
> > -----Original Message-----
> > From: condor-users-bounces@xxxxxxxxxxx
> > [mailto:condor-users-bounces@xxxxxxxxxxx] On Behalf Of Matt Hope
> > Sent: Thursday, January 19, 2006 3:01
> > To: Condor-Users Mail List
> > Subject: Re: [Condor-users] Checkpointing in vanilla
> >
> > On 1/19/06, Thomas Materna <materna@xxxxxxxxxxxxx> wrote:
> > > There is no checkpointing in vanilla universe. That is
> > precisely the
> > > main difference between the vanilla universe and the
> > standard universe!
> >
> > That is not strictly speaking true.
> >
> > There is no access by default to the automatic checkpointing
> > provided by the relinking in condor_compile.
> > Even in the vanilla universe you can make use of the condor
> > supplied checkpointing code (see previous posts or the manual
> > for this).
>
> Ok, I see what you mean, but to do that, you have to modify the code, no? If
> Junaid insists on vanilla, it is probably because he can't recompile the
> code. If he could, it would be easier to just re-link with condor_compile
> and use the standard universe.

I'm sure for most people this is the case - I just wanted to make sure
any readers understood that checkpointing (both assisted and by
whatever external means you desire) is feasible in the vanilla
universe.

There are certain cases where taking explicit control of checkpointing
but still using the condor libs to do it can be a good thing - if
perhaps you can only checkpoint at particular moments, or that you
know the cost of checkpointing and data transfer is not worth it based
on the progress made since the last one...

Matt