Intel Software Adrenaline

Programming CodeBook

Issue link:

Contents of this Issue


Page 2 of 16

Introduction 3 WELCOME As core counts and vector widths increase, programmers are challenged to take advantage of all the resulting processing power. This CodeBook explores some of the optimization techniques and models that can help developers create code that utilizes these capabilities for better performance. With Intel® software development tools, applications can scale from few cores to many cores with consistent models, languages, tools, and techniques—whether developing for Intel® Core™ or Intel® Xeon® processors, or the Intel® Xeon Phi™ coprocessor. "For James Reinders, the path to successful parallel programming is simple and achievable" Understanding how to restructure code to expose more parallelism is critically important to enable the best performance on processors, GPUs, or coprocessors. For James Reinders, Intel's director of parallel programming evangelism, the path to successful parallel programming is simple and achievable: "program with lots of threads that use vectors, while using preferred programming languages and parallelism models." This restructuring itself will generally yield benefits. Reinders emphasizes the value of common programming languages, models, and tools that work across processors and coprocessors—so developers do not have to rewrite code in order to scale their applications. Here, we'll take a brief look at some of the possibilities when using Intel processors and coprocessors. Coprocessors extend the scaling capabilities of Intel Xeon processor-based systems by exploiting additional processor vector capabilities or memory bandwidth. Explore Intel Xeon Phi coprocessor peak 1800 Discover the potential for your applications with in-depth presentations on parallel programming. 1600 1200 Coding the Future (video) » 1000 Sam 200 roce ssor sp 400 Intel Xeon coprocessor peak erf. 800 ple p Performance 1400 600 Free Webinar Series » s es ple c pro s or rf. pe Books m Sa Parallelism expert James' Reinders' blog » Intel® Xeon Phi™ Coprocessor High Performance Programming » 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 0 Threads Structured Parallel Programming » 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