ENERGY | WIRELESS | NANOTECH | MEMS | OPTICS | QUANTUM | 3D | CHIPS | ALGORITHMS

Monday, September 24, 2012

#CHIPS: "How Xeon Phi Stacks Up to GPUs"

The big question in parallel processing today is whether the Xeon Phi's 50+ x86 cores can stack up to the hundreds of thread processors on Tesla GPUs. The jury is still out, since the Xeon Phi will not be shipping until this fall, but early energy efficiency tests cited by Intel at the recent Hot Chips conference, indicate that Top500 results favor Xeon Phi (see figure below): R. Colin Johnson


Here is what Go-Parallel says about Xeon Phi versus GPU: Xeon Phi lead architect George Chrysos presented comparisons between using Xeon Phi co-processors instead of graphics-processor units (GPUs) at the recent Hot Chips conference. According to the Top500 Super Computer Sites ranking, Intel’s many-integrated core (MIC) architecture not only outperformed the two top GPU-based supercomputers on the most recent Top500 list, but was also “greener” by virtue of providing more performance-per-Watt.
Further Reading