Right up front, I am going to tell you that P-states are irrelevant, meaning they will not impact the performance of your HPC application. Nevertheless, they are important to your application in a more roundabout way. Since most of you belong to a group of untrusting and always questioning skeptics (i.e. engineers and scientists), I am going to go through the unnecessary exercise of justifying my claim.
This is the first article in a series of articles about High Performance Computing with the Intel Xeon Phi. The Intel Xeon Phi is the first commercial product of Intel to incorporate the Many Integrated Core architecture. In this article I will present the basics of the Xeon Phi architecture, the programming models and what we can do to measure the performance in cycles for micro benchmarks.
Intel(R) Composer XE 2013 Beta registration for Intel(R) Xeon Phi(tm) Coprocessor
Please read this letter, as it is explains in detail what you can expect as a result of using this Beta product. The instructions below will also direct you to a survey that will result in you obtaining a license for these products, which give you access to download the bits, and get support.
Subject: Invitation to join the Intel(R) Composer XE 2013 SP1 (compilers and libraries) beta program.
The Windows beta program for Intel(R) Xeon Phi(tm) coprocessor is no longer in effect as of September 2013. Intel now officially supports Microsoft* Windows with Intel(R) Xeon Phi(tm) coprocessor. Please visit http://software.intel.com/xeonphi and see the Quick Start guide link on the front page, which will have all of the details on how you can run a production environment with this Operating System and Intel(R) Xeon Phi(tm) Coprocessor
This is a set of labs we taught during past workshops, intended to cover more advanced concepts. These are written so that you should be able to guide yourself. The labs are available are both in C/C++ and Fortran.
Before you attempt to run these labs, make sure your environment is properly set up.
Beta Windows* enabled Drivers and Development Tools for Intel(R) Xeon Phi(tm) Coprocessor
*** UPDATE September 2013 - PLEASE READ ***
This Beta program is officially closed, as production versions of the software drivers, and supporting Intel(R) compilers are now available. Here is where to get each:
OpenMP on the Intel® Xeon Phi™ coprocessor performs as well as on Intel® Xeon processors. However, the slower clock on the Intel Xeon Phi coprocessor and the sheer number of threads accentuates OpenMP overhead. In most cases, the problem is either load imbalance or a significant amount of serial execution and is rarely the overhead itself.
Let’s take a look at the following Intel Vtune screenshot.
Cross-compilation for Intel® Xeon Phi™ Coprocessor with CMake
CMake is a cross-platform, open-source build system. A special file named “toolchain” is required for cross-compilation, and this file should define all tools (compiler, linker, libraries, etc.) needed for building an application. To invoke Cmake with the toolchain file, use the following command-line option: