A Remote Field-Programmable Custom Computing Machine Accelerator

2019 IEEE 9th Annual Computing and Communication Workshop and Conference (CCWC)(2019)

引用 3|浏览7
暂无评分
摘要
In this paper, we present a Java-based environment to access Field-Programmable Custom Computing Ma-chine (FCCM) accelerators from a remote site. We developed a hardware/software interface which supports FPGAs connected to a server. This remote environment allows users to take full advantage of FCCM accelerators by accessing FCCMs in a remote server from a client machine. Rather than providing a comprehensive development environment for FPGA synthesis, placement and routing, users simply connect to a server that runs continuously like a mail server. We present results that demonstrate the merits of our remote FCCM. In our experiments, we executed the FCCM implementation of the Taylor Series on a remote server machine and transmitted data to be processed from a client machine. Our experiments show that the execution time for the remote FCCM is not constrained by the specialized processor, but rather by the network bandwidth. On a network with 300 Mbps upload speed, and 900 Mbps download speed, execution of an application on the remote FCCM with our remote framework, with two clients and two servers, can achieve 10X speedup over the execution of a software implementation of the same application on the local machine.
更多
查看译文
关键词
Field programmable gate arrays,Servers,Hardware,Graphical user interfaces,Task analysis,Taylor series,Memory architecture
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要