Intel Software Adrenaline

Programming CodeBook

Issue link:

Contents of this Issue


Page 3 of 16

Intel® Processor and Coprocessor Overview 4 INTEL® PROCESSOR AND COPROCESSOR OVERVIEW Instruction Decoder Scalar Unit Vector Unit UV Scalar Registers Vector Registers Programs that utilize multicore processors and many-core coprocessors have a wide variety of options to meet varying needs. These options fully utilize existing widely adopted solutions, such as C, C++, Fortran, OpenMP*, MPI, and Intel® Threading Building Blocks (Intel® TBB). They are rapidly driving the development of additional emerging standards such as OpenCL*, as well as open entrants such as Intel® Cilk™ Plus. Each Intel Xeon processor and Intel Xeon Phi coprocessor core is a fully functional, multithread execution unit. Intel Xeon Phi supports a variety of programming models: 32K L1 I-cache 32K L1 D-cache • Run as an accelerator for offloaded host computation 512K L2 cache • Run as a native or MPI* compute node via IP or OFED • Run in Symmetric Mode using MPI* Ring • Run from a Windows* host machine For in-depth coverage, view the webinar: Intel® Xeon® Processors and Xeon® Phi™ Coprocessors—Introduction to High Performance Application Development for Multicore and Manycore » Extend capabilities with key features Use familiar Intel development environment • Scalar unit based on Intel® Pentium® processor family • 64-bit data path • 4 hardware threads per core • 32 512-bit vector registers per context • Fully pipelined instruction decoder • Increases resource utilization • Intel support for GDB on Intel Xeon Phi • Faster hosting on serial code • Intel® Performance Libraries (e.g. Intel® Math Kernel Library) • Intel® VTune™ Amplifier XE for performance analysis • Intel® Xeon Phi™ Processors Optimized for highly parallel applications • Use for applications that are highly parallel and can benefit from wide vectors • Based on Intel® Many Integrated Core Architecture (Intel® MIC) (up to 61 cores) • Utilizes standards-based programming languages and models Intel® MPI Library support for Intel Xeon Phi • Intel® Xeon® Processors Most commonly used parallel processor • Parallel processor supports a wide variety of applications • Provides accelerated parallel and serial performance • Offers industry-leading performance per core Intel Cilk™ Plus • Tips Intel® Threading Building Blocks (Intel® TBB) • Intel® Xeon Phi™ Coprocessor Quick Start Guide (PDF) » Intel® Composer: C, C++, and Fortran Compilers • Comprehensive resources for manycore developers » OpenMP* 4.0 • Introduction to the Intel® Xeon Phi™ Coprocessor (PDF) » Parallel programming models • Explore Standard runtime libraries, including pthreads For more information regarding performance and optimization choices in Intel® software products, visit: Privacy | Terms of Use | Learn More

Articles in this issue

Links on this page

view archives of Intel Software Adrenaline - Programming CodeBook