I had the pleasure of attending the dedication for the new HPC at the Naval Postgraduate school today.

NPS High Performance Computing Center Sun Microsystems Blade Supercomputer “Hamming” Dedication

January 30, 2009

2:00 PM

The Naval Postgraduate School is located just down the road from us here at Fleet Numerical. As such we deal with each other a bit, not nearly as much as I would like to see, but some.

NPS has put together a 4 frame, 1152 core Sun blade system with a performance of 10.7 Teraflops That’s 10.7 Trillion mathematical operations per second. Not too shabby.

The system is named after Dr. Richard Hamming, who was a Professor Emeritus at the Postgraduate school after his 30 or so years with Bell Labs. He’s famous for his communications algorithms and there is even an IEEE medal given out each year in his name.

We had the pleasure of meeting Mrs Hamming. Dr Hamming passed away in 1998 at the age of 82. I didn’t know who the little old lady with the walker was until they announced her. She waved to everyone with a shy smile when she was introduced.

There were several professors who spoke about research they planned on doing with the new system. One professor of physics spoke of taking a model simulation from a months run time to a single day.  A high speed inpact simulation. Another spoke of taking a Computational Fluid Dynamics model of a jet engine from inlet model, first stage model,…, outlet model, to a single complete airflow model through the engine from inlet to outlet.

Kudo’s out to the folks who made this possible.

  • Dr Christine Cermak- VP Information Resources
  • Dan Oliver – NPS President
  • Leanard Ferrari – NPS Provost
  • Bill Vass-  Sun Microsystems Federal President
  • Dr. Jeff Haferman – Technical Manager -HPC
  • The staff and others involved.

It’s easy for those of us in the field to forget what this kind of computational power can mean to a university or school environment. My first thought upon hearing about the size of this system was, “What’s the big deal?”. I’m spoiled in that I’ve had the pleasure of playing with some really big systems over the years. This is a really big deal to folks like those at NPS who have only had small clusters to play with before this. I’m hoping the system is hammered on by all the different professors and students. The busier the better. All my best and good luck.

Happy number crunching.

UPDATE: Dr. Jeff Haferman dropped me an email with a couple of corrections. I’ve corrected. (-: Thanks for the heads up Jeff.