From RGB
- benchmarks.  RGB submitted two bench mark proposals. Challenge 1 essentially continues the Dirac bench marking exercise
(does not use GPU code). Challenge 2 is focussed on showing that a uniform box with a bit of turbulence can be integrated on
a GPU.  The suggestion that we offer to do a version of this on ARM+Nvidia .. RGB will discuss with Andy Turner.  Hopefully
the effort to do this comes from Andy Turner’s working group. 
- contracts. The legal dept have finally sorted out how they will deal with commercial partners. This is being handled by
Rachael Matthews, who has contacted partners to assure them of progress soon. We discussed what we might expect to gain from
the interaction with the partners.
- CiSE paper. Matthieu and I are writing an article for the Computing in Science and Engineering magazine. This will be high
level and light touch. We will not include too much about current progress, rather focussing on the challenges and
describing the project. We discussed some pictures that we could use.
- June 8th meeting. 9.30-12.30.  Please let me know if you are able to attend (even if in part). We discussed how we might
handle the meeting, being realistic about the chances of Phase 1b success if we “go it alone”. Possibilities are to work
with the molecular dynamics community since the numerical methods are similar. A better approach might be to focus on the
underlying task, GPU and asynchronous communication technologies.

Quick update from others.
Edo - we briefly discussed how we might move forward on scaling tests. 

Abouzied - explained some of the compilation issues. If these are fixed then he’s optimistic about being able to have a
demonstration program that makes use of the GPU fairly soon.

Peter - is exacerbated with getting reproducible runs so that he can compare the gains down to different communication
strategies reliably. We discussed why the task frame work made this inherently difficult. 

Next meeting 21st. RGB is away on the 14th … moving the meeting a week later would let us focus on the outcome of the
cross-working group discussion.