The End of Computer Performance Modeling?

I haven't had time to digest all the details, but there's been a big song and dance this week about a supercomputer program doing in a day what took scientific minds centuries to accomplish: extrapolating Newton's laws of motion from the recorded motion of a pendulum. This diagram says it all; phenomenon in, model out:

[Source: Wired magazine]

From another perspective, this is also the holy grail of computer performance analysis: convert monitored performance data directly into performance models, feed those predictions or trends back to the computer and let it tune itself (so we can all go home).

Poor Scalability on Multicore Supercomputers

Guerrilla grad Paul P. sent me another gem in which Sandia scientists discover that more core processors don't produce more parallelism on their supercomputer applications:

"16 multicores perform barely as well as 2 for complex applications."

Worldwide Supercomputer Ratings

Interesting visualization of worldwide supercomputer performance. These Flash bubble-charts seem to be de rigueur for the NYT now. Bubble diameters are proportional to their TFLOPS rating and the location of each bubble cluster is topologically correct with respect to geographical location, but not by Euclidean distance; which is probably why it wasn't superimposed on a map.

The breakdown of these top-100 machines by processor family (not shown there) looks like this:

1. Intel: 75.6%
   
2. IBM: 12%
   
3. AMD: 12%
   
4. NEC: 0.2%
   
5. SPARC: 0.2%

However, the number 1 machine (at 1.1 petaFLOPs) is based on the IBM Cell processor.

Preposterously Parallel iPods

Here's a question: How many FLOPs in an iPod?

Climatologists, up the road here at LBL, claim that a supercomputer using about 20 million embedded microprocessors, such as those found in cellphones, iPods, and other consumer electronic devices, would deliver useful climate simulation results at a cost of $75 million to construct. A comparable supercomputer could cost closer to $1 billion. Based on a recent post, I'd be wanting to see the front-end compiler system that can upload 20 million processors.