Intel is still taking the MIC

Intel has slammed comments by Nvidia’s Jen-Hsun Huang, CEO who claimed that GPU technology was an essential ingredient on the path to reaching exascale computing within a 20MW power envelope.

Chipzilla is peddling its own version of parallel architecture which uses Many Integrated Cores (MIC), which it says will be easier for programmers to use and for the industry to scale.

James Reinders, and HPC software specialist at Intel told EETimes that Intel has managed to build a very general purpose device, one that’s designed for parallelism.

While Nvidia’s idea of sticking acclerators on to supercomputers, especially GPUs, that wasn’t necessarily the best approach.

Reinders said that it was much better to go for widely used x86 cores which have been designed for data parallelism and which are much more programmable and “even more exciting on the performance side.”

Intel was not dedicating part of its design or performance to graphics. His argument is that it is better to use a x86 device because you can do anything on it programming wise.

While 50 cores would not be “a snap to use” for everyone, people were usually “amazed” at how easy it was to work out how to program.

Reinders had a nice chat about it all here