首页> 外文会议>Global Smart Industry Conference >Associative Co-processor on the Basis of Programmable Logical Integrated Circuits for Special Purpose Computer Systems
【24h】

Associative Co-processor on the Basis of Programmable Logical Integrated Circuits for Special Purpose Computer Systems

机译:基于专用计算机系统的可编程逻辑集成电路的关联协作处理器

获取原文

摘要

An option for implementing a co-processor for special-purpose computer systems on modern hardware components is offered in the article. The aim of the paper is to create an associative co-processor, based on the FPGA base for high-performance multiprocessor systems, the ones performing associative functions and data storage functions. Retrieval and sort operations are widely used both in users' and system programs. However, these operations are the most resource-intensive and time-consuming during the traditional implementation, when the necessary information is read in the following order: from the main memory of the computer system to the processor performing the corresponding operation with them. An associative co-processor, connected to a PCI Express bus of a specialized system that implements sort, retrieval, and "moreless" comparison operations immediately on 32 words, that are preloaded into associative memory, are considered in the article. The usage of the VHDL - hardware description language - gives the opportunity to obtain a flexible project, which can be easily corrected while checking the device operation. The operation study of the proposed device was carried out in the program Web pack ISE from Xilinx - one of the leading developers of FPGA. The efficiency of the proposed associative co-processor application is provided by performing time-consuming data retrieval, sort and comparison operations directly by the coprocessor. It offloads the processor and improves the performance of the entire computer system. The main results obtained can be applied in various retrieval systems for various purposes: database servers, in the systems, used at railway stations, airports and for the efficient and rapid implementation of retrieval functions in operating systems.
机译:在文章中提供了在现代硬件组件上实现用于专用计算机系统的协处理器的选项。本文的目的是基于用于高性能多处理器系统的FPGA基础,创建一个关联协处理器,该基础是执行关联函数和数据存储功能的FPGA基础。检索和排序操作广泛使用用户和系统程序。然而,当按照以下顺序读取必要的信息时,这些操作是传统实现期间的最资格密集型和耗时的耗费量:从计算机系统的主存储器到处理器与它们执行相应的操作。关联协处理器,连接到专门系统的PCI Express总线,其在文章中考虑了在32个单词上立即实现排序,检索和“无用”比较操作的分类,检索和“无用”比较操作。 VHDL - 硬件描述语言的使用 - 提供了获取灵活项目的机会,在检查设备操作时可以轻松校正。所提出的装置的操作研究在来自Xilinx的程序网包ISE中进行 - FPGA的领先显影剂之一。通过执行耗时的数据检索,分类和比较操作,提供了所提出的关联协会共处理器应用程序的效率。它卸载了处理器并提高了整个计算机系统的性能。获得的主要结果可以在各种检索系统中应用于各种目的:数据库服务器,在铁路站,机场使用的系统中,机场和操作系统中的检索功能的高效和快速实施。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号