On May 20 & 21st, I have participated in the Matlab High Productivity Computing Advisory Board Meeting in Boston along with Dr Dan Warner of Clemson University, Mathematics Department.

The only 'Non Mathworks Inc.' presentation of the meeting was to be the report by Dr Mike Griffiths of Sheffield Uninversity and myself. It was based mainly on an earlier e-science report with more recent material added by Mike.

Mike's flight was cancelled at the last minute due to a technical fault at Manchester, England. Subsequently he could not attend the meeting in time. I ended up having to give the presentation at the last minute! I did manage to slip in a slide about the Palmetto Cluster at Clemson and talked about great things that are taking shape here in Clemson. All the slides of our presentation are available here.

Unfortunately none of the other slides, talks or any other material of this meeting will be made public. Mathwork Inc. has asked participants to keep to a confidentiality agreement about the discussions that has taken place.

Needless to say, all those who attended were Matlab experts in their own fields and were eager to contribute to the shaping of 'parallel programming' constructs in the next generation of this software. To my surprise military applications were well represented.

In order to keep within the confidentiality agreements, I shall not be able to report on the specific details of the discussions of the panel but I can perhaps make a general comment that the sticky issues of simultaneous parallelism within and outside the box have been tackled head-on with some good ideas of introducing these constructs into Matlab with minimum confusion to the user. It is likely that the implementation of many 'if not all' of these ideas will find their way into version 2009a.

Here are a few snaps, taken mainly by me during this meeting.

---

Day-2 Agenda	Break-Time	Jos Martin	Dan & Me

Author : E-mail : Last updated :

Deniz Savas, CITI. dsavas@clemson.edu