While there have been improvements in the performance of general purpose CPUs, the major gains for HPC type applications have come due to the productization of computing hardware that enables the application to run more threads than before. Manycore devices like the Intel Xeon Phi processor have given developers new options on how to design and implement applications that can run hundreds of times faster than traditional techniques.
As a bootable device, the Intel Xeon Phi is a new generation of processor that allows multithreaded applications to reside on the processor itself, without having to transfer data back and forth to a traditional host. In this way, the Intel Xeon Phi is now a fully functional processor with significantly more processing cores available than ever before. This allows for the entire application to be run on the newly designed processor. In addition, and just as important, is that cores can communicate with the network without having to go through a main CPU.
With this new design that can run both a single core, single process application to one that can run across tens of cores, more applications should show performance improvements over previous generations of devices. Work and thought will still have to go into optimizing or even taking advantage of the gain in computing power, but this new design makes it much easier.
With up to 72 out-of-order cores, the new Intel Xeon Phi processor delivers over 3 teraFLOPS (floating-point operations per second) of double-precision peak while providing 3.5 times higher performance per watt than the previous generation. As a bootable CPU with integrated architecture, the Intel Xeon Phi processor eliminates PCIe* bottlenecks, includes on-package high-bandwidth memory, and available integrated Intel Omni-Path fabric architecture to deliver fast, low-latency performance.