Tiago Cogumbreiro <cogumbre...@di.fc.ul.pt> wrote on 05/29/2014 11:52:05
AM:
>
> I am having some difficulties using the x10-benchmark-2.4.3.
>
> I am struggling to know which parameters are valid and how to
> interpret the results printed in the screen. When I try to input
> parameters different than the ones in the bundled scripts, the
> benchmark just hangs/crashes.
>
> What is this benchmark suite? Is there any documentation? Is it the
> one described in the paper "X10 and APGAS at Petascale"?
Hi,
These are the same 8 benchmarks from the "X10 and APGAS at Petascale"
PPoPP'14 paper, but updated to X10 2.4.3. (The experimental results in the
PPoPP paper were on X10 2.2.3).
We could certainly do a better job of documenting the constraints on
the input parameters. In many cases there is a block comment at the top of
the source file that gives some hint. In general, the focus of these
benchmarks was looking at scaling to a large number of nodes, so there is a
bias towards the code assuming specific combinations of parameters and
relationships between them.
If there are specific benchmarks that are giving you trouble, we can
probably provide more guidance.
--dave
------------------------------------------------------------------------------
Time is money. Stop wasting it! Get your web API in 5 minutes.
www.restlet.com/download
http://p.sf.net/sfu/restlet
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users
