论文笔记（6.FCCM.2018）

最新推荐文章于 2024-07-07 14:15:04 发布

Jason_1141

最新推荐文章于 2024-07-07 14:15:04 发布

阅读量298

点赞数

本文链接：https://blog.csdn.net/qq_40884849/article/details/113361340

版权

Information

Paper:Hot & Spicy: Improving Productivity with Python and HLS for FPGAs
Author: Sam Skalicky
Key words: FPGA,Python

Backgrounds

Today’s HLS tools still rely on C/C++ which is suitable for conventional embedded platforms, but do not yet support more modern programming languages.
This paper tries to solve the following qusetion:

“Can we support accelerating portions of a pure Python application in FPGAs?”

在这里插入图片描述

Work

This paper presents Hot & Spicy, a tool suite for integrating FPGA accelerators in Python applications,which includes:

Open-source Python to HLS-suitable C translator
Open-source tools to support integrating FPGA-based
accelerators into Python applications
Analysis of overheads of using Python to control FPGA-based accelerators compared to C/C++
Experimental results showing realizable speedup for Python implemented algorithm accelerated via source-to-source translation to C and implemented via HLS