Man thats a beast of a machine - each of the CPUs have 8b of cache, not to mention that you get a terra flop of double precision.
Now if only I had some use for this :( (inspiration welcome, I am writing about GPU programming right now).
Now if only I had some use for this :( (inspiration welcome, I am writing about GPU programming right now).
What is the mapreduce of CUDA going to be?