A few months ago I ran a piece entitled Fading Glory, where I made the following observation re the place where I’m working on my PhD in Computational Engineering:
One of the things my advisor oriented us about were the substantial computer capabilities that the SimCentre has. “Substantial” is a relative term; at one point we were in the top 500 computers in the world for power, but he chronicled our falling ranking which eventually left us “off the charts”…What has happened to the SimCentre was not that the computer cluster there had deteriorated; it has just not kept up with the ever-larger supercomputers coming on-line out there.
Forecast models require some serious computational horsepower, which can only be supplied by supercomputers. The ECMWF, for example, utilizes an IBM system capable of over 600 teraflops that ranks among the most powerful in the world, and it’s used specifically for medium-range models That, fundamentally, is the reason their model frequently outperforms the American one. The US National Weather Service’s modeling center runs a diversity of short-, medium-, and long-term models, all on a much smaller supercomputer. The National Weather Service has to do more with less.
I’ve lamented the way we’ve allowed our transportation infrastructure to deteriorate, but that also applies to other parts of our infrastructure, which includes simulation (as opposed to the statistical extrapolations which are so popular). At the root the United States, for all the blessings it has been endowed with, has a broad-based “fading glory” problem based on its dysfunctional political system, and that in turn is based on its dysfunctional society.
People in the business routinely talk about computational routines being “expensive” but in reality it’s nothing compared to the damage of an unexpected hurricane. Of course, we might have some more breathing room if we wouldn’t overdevelop our coastal areas, but that’s another post.