There's a group of computer scientists at the University of California, Berkeley that have been looking at the future of parallel computing and put out a report a couple of months ago; I finally had an opportunity to read this, and I found it to be a very though-provoking look at the prospects and research priorities in computing hardware and software involving or relying on parallel computing. While parts of this report are quite somewhat technical, I think that many CNI-announce readers will find at at least the first sections to be interesting reading.

The group has a site at:

and the report is linked from this site, or available directly at:

http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-183.pdf

Clifford Lynch Director, CNI