arrow
Return

A scala based framework for developing acceleration systems with FPGAs

delete2019-09-01
delete9
PRE
AI
Y
Yanqiang Liu
Y
Yao Li
戚正伟 (Zhengwei Qi) *
H
Haibing Guan
DOI:10.1016/j.sysarc.2019.08.001delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Field-Programmable Gate Arrays (FPGAs) in heterogeneous computing have been attracting more and more attention. Development of acceleration systems based on FPGAs involves the cooperation of both hardware and software developers. However, although most hardware acceleration systems are motivated by software developers, software developers are difficult to participant in the system building because of the steep learning curve of the hardware concept and design tools. Moreover, due to the complexity of hardware-software integration in traditional ways, end-to-end performance is hard to evaluate before trivial engineering effort. To address these concerns, we propose an open-source Domain-Specific Language (DSL) based framework called VeriScala(1) to support hardware defining in a high-level language, programmatical testing, and rapid acceleration system deploying. By adopting DSL embedded in Scala language, we introduce modern software development concepts into hardware design and provide a familiar environment to software developers. And by building a stack of middle layers from hardware to software, we provide reduced hardware abstraction and communication interface to facilitate both software and hardware development and system deployment. Through the evaluation of some basic components and real-world demos, we show that VeriScala provides a practical approach to rapid prototyping of hardware acceleration systems.
Keywords:
FPGA
DSL
EDA
Scala
AI Summary

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.

Journal

Journal of Systems Architecture cover
Journal of Systems Architecture
IF:
4.1
Papers:
3.0K
Citations:
4.2K

Organization

S
shanghai jiao tong university
Scholars:
15.6W
Papers: 11.6W
Citations: 159