[Users] Tracer particles: Processor distribution of array

[Roland Haas]

Hello Jens,

there is no support (that I would know of) for having per-variable
control how the array is distributed though. Ie the choice made in the
parameters affects all variables. This could be changed with changes in
Carpet, at least for grid-arrays (which are the only ones you are
concerned with). Note that this split is fixed in time for grid arrays
at this point, it cannot be adjusted. Making it adjustable would
require changes inside of Carpet (making the grid-array more similar
to a grid-function).

I am assuming you'd like to use grid arrays to have access to
checkpointing / recovery? Otherwise you could use just regular vectors
and resize them yourself as required.

Yours,
Roland

> Jens
> 
> The "DISTRIB=..." options only determine on which process the particles are
> held. This is a purely computational property and is independent of any
> physics. You cannot change this without updating the time integration
> algorithm.
> 
> The particles are set up via some kind of initial condition, and there
> might be a run-time parameter that you can set to modify this.
> 
> -erik
> 
> 
> On Wed, May 10, 2017 at 8:42 AM, Jens Mahlmann <jensmahl at alumni.uv.es>
> wrote:
> 
> > Dear all,
> >
> > I am currently working on the implementation of tracer particles into an
> > evolution code. As for this, my orientation is the particle implementation
> > of GRHydro. The number of particles is, in this case, is distributed
> > equally across the processors by the DISTRIB=DEFAULT option. Is there any
> > experience out there with distributing arrays according to a transfer
> > function (e.g., some sort of density), such that higher density regions
> > have more traced particles? In the documentation I only found the
> > DISTRIB=CONSTANT option - which would not serve the purpose.
> >
> > Thank you very much and my best!
> >
> > Jens
> >
> > _______________________________________________
> > Users mailing list
> > Users at einsteintoolkit.org
> > http://lists.einsteintoolkit.org/mailman/listinfo/users
> >
> >  
> 
> 



-- 
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://pgp.mit.edu .
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
Url : http://lists.einsteintoolkit.org/pipermail/users/attachments/20170510/c6585ea8/attachment-0001.bin