Niagara2 for HPC ?
As most of us will have already know that Niagara take on the school that we should have simple pipeline and with multicore and multithread. The first Niagara processor has only one float-point unit shared among all the cores and threads. In Niagara 2, the number of floating units is increased to 1 per core which make is attractive and interesting for some of the floating point computation intensive application. While the chip’s peak floating point performance of 11,2 GFlop/s quite moderate, the high potential of the Niagara 2 is revealed, when many threads are active and the high memory bandwidth of some 60 GB/s (theoretically) can be exploited - a frequent bottleneck of standard architectures when executing technical applications.
Take at look at the result at http://www.rz.rwth-aachen.de/ca/k/raw/?lang=en
