摘要
随着四核微机走向市场和八十核处理器在实验室研制成功,多核正引领软件研发发生基础性变化。开发人员需要在代码中添加线程来利用系统所提供的多个内核,从而提升PC应用软件的功能和性能。文中探讨在多核微机上进行并行计算的实现技术。介绍了共享存储系统并行编程接口OpenMP的模型、指令和库函数,以及Intel C++编译器9.1和Microsoft Visual Studio 2005等对OpenMP的支持;着重探讨了二维离散快速傅里叶变换并行算法的设计、实现与优化技术;展望了高性能并行计算软构件库的开发前景。
Software development will receive a foundational innovation by reason of multi - core technology, along with quad- core PC coming into the market and invention of 80 - core in the laboratory. Programmers add threads into codes to make full use of the new processor, that improves both function and performance of Internet applications. Originates from parallel programming on multi - core com- puters. First introduced OpenMP standard which is an application programming interface(API) on parallel programming model of shared -memory,and gave a quick overview about a set of compiler directives and a library of support functions. Using the OpenMP programs requires an OpenMP- compatible compiler and thread- safe libraries, both Intel C + + compiler 9.1 and Microsoft Visual Studio 2005 are perfect choice. Then studied two-dimensional discrete fast Fourier transform(FFT), focused on parallel program design, realization and optimization technology. Finally, thought that high performance parallel computing software component library must be a perfect exploitation field in the further future.
出处
《计算机技术与发展》
2007年第10期87-91,共5页
Computer Technology and Development