Fine grain parallelism on a MIMD machine using FPGAs

[1993] Proceedings IEEE Workshop on FPGAs for Custom Computing Machines Pub Date : 1993-04-05 DOI:10.1109/FPGA.1993.279485

Frédéric Raimbault, Dominique Lavenier, Stéphane Rubini, Bernard Pottier

引用次数: 32

Abstract

The article presents the use of an FPGA chip (Xilinx 3090) to set up a fast systolic communication agent on a linear asynchronous network of transputer processors; the machine is called ArMen. The authors' work relies on the systolic programming environment ReLaCS, a close cousin to the C programming language. ReLaCS provides synchronous communication operators to simplify the programming of data transfers that occur in systolic algorithms. The ReLaCS compiler generates C programs that perform the computation process and the data management process of a systolic network.<>

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

使用fpga的MIMD机器上的细粒度并行性

本文介绍了利用FPGA芯片(Xilinx 3090)在一个线性异步网络上建立一个快速收缩通信代理;这台机器被称为ArMen。作者的工作依赖于收缩编程环境ReLaCS，它是C编程语言的近亲。ReLaCS提供同步通信操作符，以简化在收缩算法中发生的数据传输编程。ReLaCS编译器生成C程序，执行收缩网络的计算过程和数据管理过程

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

[1993] Proceedings IEEE Workshop on FPGAs for Custom Computing Machines

自引率

0.00%

发文量

期刊最新文献

Virtual wires: overcoming pin limitations in FPGA-based logic emulators The CM-2X: a hybrid CM-2/Xilinx prototype Hardware acceleration of divide-and-conquer paradigms: a case study Data-folding in SRAM configurable FPGAs WASMII: a data driven computer on a virtual hardware