New crop of Chinese supercomputers will feature homegrown chips.
In contrast to the previous decade, CPU clock rates are scaling slower over time due to the power constraints. However, the number of transistors per silicon area continue to increase roughly at the rate of Moore’s Law. Therefore, CPUs are being designed and built with an increasing number of cores, with each core executing one or more threads of instructions. This puts a new kind of pressure on the memory subsystem.
Yes, there is life beyond Xeons, Opterons and GPGPUs.
Top seven supercomputers make it into the petaflop club.
The tension between custom and commodity high performance computing has shaped both market approaches.
Integrated graphics threatens GPU pricing subsidies.
HotPar workshop spotlights latest work in parallelism.
Nvidia Fellow David Kirk takes a swipe at Intel’s heterogeneous computing plans.
The CPU-GPU performance wars bring out the pitfalls of benchmarking.
AMD predicts bigger role for GPUs in HPC servers.