3 ms·
I really can't imagine this has any impact on any chinese intelligence agency whatsoever. Tianhe-2 is already the #1 supercomputer - what possible difference c
by emn13 12y ago
I really can't imagine this has any impact on any chinese intelligence agency whatsoever. Tianhe-2 is already the #1 supercomputer - what possible difference could a minor (it's not an order of magnituge faster, and software doesn't scale perfectly anyhow) upgrade make?
- Kalium 12y agoIt cuts off one avenue by which a foreign country might gain useful information from within American companies. China in particular is notorious for their industrial espionage.
- kuschku 12y agoThat’s actually a hilarious statement, considering how notorious the US is for their industrial espionage. Both china and the US are equally bad in that situation, and both equally try to fully observe every single step any of their people does.
- outworlder 12y agoNot the same, though. The US does quite a lot of original research. Industrial espionage in this context is more about predicting the adversary's moves, not stealing their IP. The 'stealing' part takes on many forms. One of them is: http://thediplomat.com/2013/11/how-china-plans-to-use-the-su-35/ http://thediplomat.com/2013/11/how-china-plans-to-use-the-su... > Russia understandably became upset when its star export appeared as an indigenously produced J-11 in China – without a licensing agreement. In short, China has a pretty large technological gap, and does anything necessary to shorten the gap as quickly as possible. And that's the government. There's no indication that private Chinese companies are any better in this regard. In fact, you can even find 'counterfeit' cars. One wonders why US companies outsource their manufacturing there. Their technology will be 're purposed', without any need of industrial espionage, by the standard definition of it.
- semi-extrinsic 12y agoIt's funny, because technically Tianhe-2 is the world's fastest supercomputer based on Linpack, but most of that comes from the Xeon Phi accelerators, which to my knowledge has been a total flop on the HPC scene. It's almost like a brilliant evil plan to make the Chinese chase a futureless accelerator platform.