Server

英特尔® 至强® 处理器和英特尔® 至强融核™ 协处理器的浮点算法差异

下载文章

下载英特尔® 至强® 处理器和英特尔® 至强融核™ 协处理器的浮点算法差异 [PDF 120KB]

本文介绍英特尔® 编译器使用的浮点模型及其在英特尔® 至强® 处理器场景中的应用“英特尔® 编译器浮点运算结果的一致性”。 在正确设置的情况下,编译器生成的代码完全符合面向二进制浮点算法的 ANSI 语言标准和 IEEE-754 标准。 编译器选项支持用户合理控制性能优化、精确度、结果再现性和标准合规性。

  • Server
  • Intel® Xeon Processor
  • Intel® Xeon Phi™ Coprocessor
  • Floating Point Arithmetic
  • Intel® Compiler
  • Intel® Composer XE 2013
  • Intel® Many Integrated Core Architecture
  • Parallel Computing
  • 面向英特尔® 至强融核™ 协处理器的 OpenCL* 设计和编程指南

    关于本文档

    本文旨在对开发面向英特尔® 至强融核™ 协处理器的高性能 OpenCL 应用所需的设计和编码指南进行介绍。 阅读本文,您可以了解英特尔至强融核协处理器架构和微架构、关键的 OpenCL 结构,以及学习如何通过高效地使用这些架构最大限度地利用英特尔至强融核协处理器硬件。 由于使用硬件的并行性对高性能应用至关重要,因此我们将为您展示如何在英特尔至强融核协处理器上改善 OpenCL 应用的并行性。 这些知识将帮助您对您的应用进行有效的设计和编程,以便在英特尔至强融核协处理器上最大限度地利用 OpenCL。

  • Server
  • OpenCL*
  • MIC
  • Intel Xeon Phi
  • PCIe bus
  • vcsource_product_oclsdk
  • vcsource_type_techarticle
  • vcsource_os_windows
  • vcsource_domain_graphics
  • 面向英特尔® 至强融核™ 协处理器的英特尔和第三方工具与库

    众多工具厂商已经宣布将提供面向英特尔(R) 集成众核架构的定制软件版本,首先是面向英特尔(R) 至强融核(tm) 协处理器。  请联系厂商详细了解英特尔(R) 至强融核(tm) 协处理器支持的版本信息。

    这一“实时”文档将随新软件的推出而更新。

  • Server
  • Development Tools
  • Intel® Many Integrated Core Architecture
  • Parallel Computing
  • Vectorization
  • 面向英特尔® 至强融核™ 协处理器的系统管理

    更新日期:2013 年 7 月 18 日

    下载


    面向英特尔® 至强融核™ 协处理器的系统管理 [PDF 597.22 KB] pre-MPSS 3.1

    Under revision! At present, the version of the system administration guide here applies only to those versions of the MPSS earlier than 3.1

    前言


    本文档概述面向英特尔® 至强融核™ 协处理器的系统管理。主要面向小规模系统管理员。本文档不能取代英特尔® 集成众核架构(英特尔® MPSS)的附带文档而是作为补充,其中包括相关提示、故障排除建议以及其它文档的指针。

    概述

    本文从系统管理员的角度研究英特尔至强融核协处理器。

  • Developers
  • Server
  • MIC
  • Intel® Xeon Phi™ Coprocessor
  • Intel® Xeon Phi™ Coprocessor System Admistration Guide
  • Profiling OpenMP* applications with Intel® VTune™ Amplifier XE

    Parallelism delivers the performance High Performance Computing (HPC) requires. The parallelism runs across several layers: super scalar, vector instructions, threading and distributed memory with message passing.    OpenMP* is a commonly used threading abstraction, especially in HPC. Many HPC applications are moving to a hybrid shared memory/distributed programming model where both OpenMP* and MPI* are used.

  • Developers
  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8
  • C/C++
  • Fortran
  • Intel® VTune™ Amplifier XE
  • Intel VTune Amplifier
  • OpenMP*
  • Development Tools
  • Optimization
  • Parallel Computing
  • Привязка потоков (affinity) в Intel® Threading Building Blocks на сопроцессоре Intel® Xeon Phi™

    Библиотека Intel® Threading Building Blocks (Intel® TBB) [1] [2] предоставляет высокоуровневые интерфейсы для написания программ, использующих параллельные вычисления. И несмотря на то, что цель этих интерфейсов скрыть от разработчика управление потоками в системе, всё же иногда встречаются задачи, требующие ручного управления. Одна из таких задач - привязка программных потоков к определённым физическим потокам (ядрам) [3].

  • Developers
  • Students
  • Linux*
  • Business Client
  • Server
  • C/C++
  • Intel® C++ Composer XE
  • Intel® Threading Building Blocks
  • Intel Threading Building Blocks (Intel TBB)
  • Intel Xeon Phi
  • Affinity
  • tbb affinity
  • cache efficiency
  • observer
  • task_scheduler_observer
  • pi
  • тбб
  • аффинити
  • кэш
  • привязка потоков
  • What’s New in Intel® Composer XE 2013 SP1

    Intel® Composer XE 2013 SP1 includes Intel® Compiler 14.0 among other components. The list below summarizes the features and enhancement highlights in Intel® Compiler 14.0 that are pertinent to those programming for Intel® Xeon Phi™ coprocessors:

    ·       Support for the new Intel® Xeon Phi™ Coprocessor, Code Named Knights Landing  

    Intel® Trace Analyzer and Collector 9.0 Beta Readme

    The Intel® Trace Analyzer and Collector is a low-overhead scalable event-tracing library with graphical analysis that reduces the time it takes an application developer to enable maximum performance of cluster applications. This Beta package is for users who develop on and build for Intel® 64 architectures on Linux* and Windows*, as well as customers running on the Intel® Xeon Phi™ coprocessor on Linux*. You must have a valid license to download, install and use this product.

  • Linux*
  • Microsoft Windows* (XP, Vista, 7)
  • Microsoft Windows* 8
  • C/C++
  • Fortran
  • Intel® Trace Analyzer and Collector
  • mpi-3
  • mpi analysis
  • profiling
  • cluster tools
  • Message Passing Interface
  • Cluster Computing
  • Symmetric Mode MPI Performance without InfiniBand*

    Symptom

    Slow host-coprocessor MPI communications in systems with no InfiniBand* HCA.  If running with I_MPI_DEBUG=2 or higher, you will see one of the following messages indicating that the TCP fabric has been selected:

    [0] MPI startup(): tcp data transfer mode
    [0] MPI startup(): shm and tcp data transfer modes
  • Linux*
  • Server
  • Intermediate
  • Intel® MPI Library
  • tips and tricks
  • Configuring Intel® Xeon Phi™ coprocessors inside a cluster
  • Intel® MPSS
  • ofed
  • Message Passing Interface
  • Intel® Many Integrated Core Architecture
  • Applying Intel® Threading Building Blocks observers for thread affinity on Intel® Xeon Phi™ coprocessors.

    In spite of the fact that the Intel® Threading Building Blocks (Intel® TBB) library [1] [2] provides high-level task based parallelism intended to hide software thread management, sometimes thread related problems arise. One of these problems is thread affinity [3]. Since thread affinity may help to optimize cache performance [3] [4] and consequently overall performance, this topic cannot be avoided.

    Subscribe to Server