As I write this, I’m headed back up to Seattle on the Amtrak – not the fastest way to move between Portland and Seattle, but (IMHO) the one which provides the highest quality experience. No security, no lines and no restrictions during take-off and landing; virtually door-to-door service between downtown Portland and Seattle and none of the hassle of traffic. Next to someone who is crocheting, several are sleeping and the rest are simply enjoying the view and the comfort.
This experience is very analogous to an Isilon cluster (compared to the other clustered systems at SuperComputing) – not the fastest, not the one which offers the most do-it-yourself flexibility – but the one that will give you the highest quality experience in terms of TCO and simplicity.
As much as I would love to imagine about a simpler time for travel, I will instead look forward – and I think what I saw at SuperComputing represents the future. Not the technology so much as the new challenges the technology is being used to address.
Anyone who attended the show had an opportunity to see Al Gore’s keynote today (Thursday) and regardless of your stance on the particular issues he called attention to, I was definitely moved by the call to action. The power of the supercomputers in use today (which will double in 18-24 months, according to Moore’s law) is being directed at problems which are extremely relevant to comfort, society and the overall survival of our species.
Right across from the Isilon booth was Oak Ridge National Laboratory, home of the world’s fastest supercomputer at approximately 1.7 petaflops. That power is being used to use to improve the efficiency of ethanol fuels at the molecular level, to analyze and understand dark matter in the universe, to better understand climate change and its implications, etc. Down the aisle was a German institute (http://www.kit.edu/) focused on delivering computer-driven cars and kitty-corner to their booth was the Japanese Atomic Energy Agency focused on improving the use of nuclear energy.
Isilon clusters are not at any of these particular examples, but are in use at a wide variety of research institutions, including San Diego Supercomputer Center, the Broad Institute of MIT and Harvard and the J. Craig Venter Institute.
At the end of the day, knowing that I’m part (in a tiny way) of the critical research aimed at solving some of our society’s greatest challenges gives me a wonderful feeling of pride.
