Date: Fri, 15 Sep 2006 13:22:52 -0400
Subject: Blue Gene supercomputer to go into full production

Dear Researcher,

On behalf of the Center for Computational Science and the Scientific 
Computing and Visualization group in the Office of Information 
Technology, we are very pleased to announce that we will put our IBM 
Blue Gene supercomputer into full production as of October 1, 2006. 
This system has a peak capability of 5.7 TFLOPS and represents a vast 
increase in our production capabilities. Our machine is ranked among 
the top 100 fastest computers in the world. 

Our Blue Gene is being used very successfully by a number of 
researchers from our community during our extended Friendly User 
period. At this time we would like to make this important resource 
available to all of our researchers. At the same time we will be making 
some small changes to our procedures for issuing computer accounts and 
will initiate this procedure for new users of the Blue Gene system. Over 
the next few weeks we will send out more details on this process.

Once the Blue Gene goes into production, we will be charging for its use 
against your research project's computer time allocation. Our machine 
comprises 1024 compute nodes. Each compute node contains two 
PowerPC processors running at 700MHz and 512 MB of memory. The 
charge rate will be 0.25 SUs per node-hour . (One SU equals one 
processor hour on the p690). Unlike the p690 and Linux clusters, the 
charge is based on the job's actual running time (i.e. wall-clock hours), 
not processor (CPU) hours. This offers a quite favorable rate for those 
jobs that can take advantage of the Blue Gene architecture, particularly 
those that can utilize both processors on a node.

For more information on the BG/L, please visit the SCV Web site at 
http://scv.bu.edu/SCV/bluegene.html. If you would like to start using the 
Blue Gene immediately, you may still apply for a Friendly User account 
following the instructions on the Web page. We will be sending out 
further information on obtaining access to the Blue Gene over the next 
few weeks.

We are most pleased to be able to bring this resource into production and 
hope you will be able to make good use of it in furthering your research.

Sincerely, 

Claudio Rebbi, Center for Computation Science
Glenn Bresnahan, Scientific Computing and Visualization