PACE A Partnership for an Advanced Computing Environment

February 18, 2010

HPC status update – 2/18/2010

Filed under: Inchworm deployment: Winter 2009 — admin @ 3:27 pm


Here is an update on the combined HPC purchase.  Essentially all equipment has been delivered.  Penguin is on-site and has begun installation and parts inventory.  Once the cluster is set up, it will undergo a series of acceptance tests, including several that require multiple days of error-free processing.  Tests will be run at the unit level, concurrently on each cluster, and cluster-wide.  The original schedule called for Penguin to run these tests at their facility.  However, Penguin and GT have agreed to perform these tests on-site to assure that all tests specified in the Agreement are completed successfully.  This also mitigates some impact on the schedule as a result of parts supply delays, which have now been resolved.  In light of this, we’ve also eliminated the need for a visit to Penguin’s factory in California.

FYI, below is the current schedule – some further adjustments may be necessary pending results of testing.  Now that the installation is underway, we’ll be sending out updates a little more frequently.

mid February (ON TARGET)

  • delivery of equipment

late February

  • setup of equipment to be used by users for “early testing”

mid/late March

  • completion of on-site testing by Penguin

late March/early April

  • OIT completes software installation
  • ready for user acceptance testing

early/mid April

  • ready for research

A reminder: If you have plans, or know of anybody who does, for additional HPC systems in the next 18 months or so, please contact me.

Powered by WordPress