[Swift-devel] Re: scheduling

[Michael Wilde]

I was thinking that mappers might be the place to look up the location of a logical file and return a physical one, but I can see how that would be hard to integrate into the scheduling model.

I'd say keep it on the idea stack, proceeding as Mihael suggest sounds great, and lets see where that takes us.

- Mike

----- "Mihael Hategan" <hategan at mcs.anl.gov> wrote:

> On Wed, 2010-03-31 at 03:43 +0530, Vipul Kumar Singh wrote:
> > Hi,
> > 
> > Going through the code i realized that we could add to the
> > checkConstraints() which returns the desired sites to the
> > getNextResource()...
> 
> You're close to what I think should be done.
> 
> My opinion is that availability of files should bias the site scores
> instead of imposing hard constraints. There would be a parameter that
> would control the strength of that bias.
> 
> >  We can add data dependencies as task constraints, the constraints
> > checker then checks the availability of data on sites based on
> > catalog, as you suggested. So the logical step would be to have the
> > support for RLS catalog in swift so it can be used to make
> scheduling
> > decisions.
> > 
> > But I am having trouble figuring out how to get the mappers and
> > catalog concept work complementary to each other i.e. what would be
> > first steps towards having a catalog of data in swift..
> 
> Right. Swift would need some form of data catalog. Perhaps a simple
> in-memory catalog would be a good first start. And before that there
> would be a need for an API specification to which various
> implementations could be plugged (e.g. in-memory, database, RLS).
> 
> Also, I would say ignore mappers for now. That is unless Mike has a
> good
> reason for doing otherwise that I'm not aware of.
> 
> Mihael
> 
> _______________________________________________
> Swift-devel mailing list
> Swift-devel at ci.uchicago.edu
> http://mail.ci.uchicago.edu/mailman/listinfo/swift-devel

-- 
Michael Wilde
Computation Institute, University of Chicago
Mathematics and Computer Science Division
Argonne National Laboratory