Why is the term "Supercomputer" being used to describe Watson? No demonstrated systems have shown anywhere near the processor or node count that actual supercomputers have (the Watson machine on Jeopardy for instance was only 90 nodes with around 2K cores). Also it uses an off the shelf interconnect (10gbit fiber) with a simple hierarchical network fabric which doesn't even approach even small supercomputers in terms of performance (which use something like Infinband or Seastar in a N-Dimensional torus interconnect topology).

While I have nothing against the technology being used for Watson. The fact is that it is not a supercomputer and the division of IBM that did make supercomputers (BlueGene) has been disbanded (with most of the key individuals leaving for other places).

I saw this poster at the conference and I was not impressed and in fact it was one of the weaker posters that I saw at the conference (it was light on details and had some of the information on the poster when talking about GPU's in general was not entirely accurate). It is really a poster that should not have been at SC at all. While it is interesting in the network sense the amount of data they can process is not anywhere close to the amount that is actually flowing through these large scale machines (up to 10 GB/sec per node) and there was no information about scaling this data collection (which would be needed at extreme scales) to obtain meaningful information to allow for tuning of network performance.

This poster should have been at a networking conference where the results would have been much more interesting to the crowd attending. Also of note, IIRC the author was using a traditional GPU programming model for computation that is not efficient for this style of computation. The speedup numbers would have been greatly improved by using a RPC style model of programming for the GPU (persistent kernel with tasking from pinned pages). However this is not something I totally fault the author for not using since it is a rather obscure programming technique for GPU's at this time.

"Supercomputing applications tend to require cores to work in concert with each other, which is why IBM, Cray, and other companies have built incredibly fast interconnects. Cycle's work with the Amazon cloud has focused on HPC workloads without that requirement." While this is cool, Can you really call something like this an HPC system if you are picking work loads that require little cross node communication? The requirement of cross node communication is pretty much the whole reason large scale HPC machines like ORNL's Titan exist at all. Wouldn't this system be classified closer to HTC because it is targeting workloads that are similar to those which would be able to run on HTC Condor pools?

The first iPhone also wasn't competing against three other smartphone platforms.

No kidding. They have the absolute WORST API documentation of any product i have ever used. Its slightly better now (in Arc 9 good luck finding documentation to do anything non-standard). However this problem does extend beyond ESRI since it seems like documentation in this entire software sector is pretty shitty (OGC is better but not great).

They do not. However there have now been questions raised as to the analysis by Musk by other papers. http://www.theatlanticwire.com/technology/2013/02/elon-musks-data-doesnt-back-his-claims-new-york-times-fakery/62149/

There are already 3 at national labs that are > 10 PFLOPS... ORNL's Titan, LLNL's BG/Q, and ANL's BG/Q. The story behind this is they want it cheaper than the crazy price tag those machines have.

I wouldn't jump the gun on this yet. While it is a really promising sign that devkits are making their way out, the post on Ouya's site has something quite concerning. "The dev consoles aren’t cheap for us to make." No one claimed they couldn't make this console. Most of the issues brought up was that people doubted that they would able to make it cheap enough to meet a $100 price tag (dev kits were $699+ each).

Couldn't you modify android to unrestrict this on some (or most) android tablets?

The WII U has 2GB of memory but only 1GB available to games (1GB is reserved for the system).

Because some (or most) new software engineers hadn't really worked real jobs prior to their new careers and believe that because the company provides food that they instantly care about your well being. This is not the case. I am not saying to go screw over your employer, but you should know not to be screwed over by them.

You sir, are not familiar with the bureaucracy they call the US government. It wouldn't surprise me if they have to spend $1 Billion to be able to store only 50 Terabytes of the data..

What makes the wii better suited for distribution of video over the Xbox 360 especially since the 360 has netflix already in its camp?