The 36-port part is the ASIC. The switch boxes have a lot more ports.

Not that it changes your argument, but you should know that NCSA has a brand new Altix.

It appears to be the latter. The spec is available here. NCSA negotiated a system with IBM, proposed it to NSF under the above linked RFP, went through a peer-reviewed awards process, negotiated an award with NSF, and started working on the delivery and other aspects with IBM and NCSA's other partners. Something went wrong in the last several months, and IBM's pull out was the result. I doubt that there is any more money to be found, and all parties knew what was asked of them in order for the project to be successful.

Have you tried it off-node?

Have you tried fitting a water block in a blade lately? How about 2000 of them? :)

Given that you can lay two of these racks back to back and then run them end to end, and that you can remove most of the regular AC equipment from your room, the amount of stuff you can get in your datacenter is the same.

It's 3% on my Citibank card here in the US. That's the normal rate for US-based banks for foreign transactions, BTW.

You don't understand how this works. You do the computation ahead of time on the supercomputer to build your reduced order model which you download onto your phone and take out into the field. Once you've downloaded the model, you don't need the supercomputer any more. You can use the phone to do computations using the reduced model as much as you like. If you get into a regime where the predicted error from the reduced order model is too high, you can go back to the supercomputer and update the model. If that happens, then you'll probably have to wait in queue again, but that's not such a big deal.

You might find this interesting.

Unless we've all got InfiniBand between us, it's not really worth it. These simulations are tightly coupled across the nodes they run on making them very sensitive to the latency and bandwidth between them.

I should have pointed out above that we measure in wall-clock time not CPU time. Most of these codes don't spend much time waiting on I/O, so the two numbers are usually close. We use wall-clock time because that is the time that the user monopolizes the nodes that are assigned to their job.

The code in question (ADCIRC) has been used for years to do hurricane storm surge simulations. It's being continuously developed for work in the Gulf of Mexico and already includes contaminant transport effects. Also, as with all things scientific, "right" is a relative quantity. The better question is whether or not useful predictions can be made that are better than what's been done so far. I think the answer to that is a resounding "Yes!" Finally, I guarantee that this event will be used by modelers to refine and improve their codes for years to come. Recent hurricances (Ike, Rita, Gustav, etc.) have been used in the very same way.

The model itself can get quite close to shore (much closer than the pixel-level resolution of that tiny map you linked to), includes wetting and drying of land regions, and has been used to predict (in both forecasting and hindcasting modes) hurricane storm surge.

The simulation is currently running on about 4k cores which is about 244 hours or 10 days worth of simulation. Each of the simulations they're running is about 10 hours in length, so this is enough for about 24 forecasts.

End users rarely pay for supercomputer time. Very rarely. There's an application process and peer review in most cases, but the time is, in the end, free.