Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
Monstertail 2b1bdb13cf | 1 year ago | |
---|---|---|
benchmark | 1 year ago | |
src | 1 year ago | |
Makefile | 1 year ago | |
description.json | 1 year ago | |
krnl_graph_stream.cfg | 1 year ago | |
qor.json | 1 year ago | |
readme.txt | 1 year ago | |
utils.mk | 1 year ago | |
xrt.ini | 1 year ago |
[环境配置]
1. vitis HLS(推荐命令行的版本,使用方式请参考xilinx的vitis官方文档)
[代码结构和运行方式]
本项目通过HLS开发在Vitis平台完成了FPGA设计,并通过端口设计提高计算单元数量、Kernel结构调整、读写策略优化、流水线设计等手段,
实现了比一开始的设计方案性能提升6倍的FPGA加速器,且该加速器性能在OGB Products和OGB Proteins数据集上达到了和CPU(Xeon(R)Silver 4214)
48线程相比×1.22到×1.99倍的加速
1. host端见src/host.cpp,主要是进行加速器的分配设置和host端的数据传输。
2.kernel端分成读和写两个cpp文件。读在src/krnl_graph_read.cpp,写在src/krnl_graph_write.cpp。控制加速器的运行逻辑。和host端的传输接口要对应。
3. 加速器配置见krnl_graph_stream.cfg,src/krnl_graph_stream里面的只是副本。
4.benchmark(包括CPU,GPU,一开始的FPGA设计)在benchmark/
[联系方式]
有问题请联系作者姚锦炜 jinwei.yao@epfl.ch//QQ:769845882。
本项目通过HLS开发在Vitis平台完成了FPGA设计,并通过端口设计提高计算单元数量、Kernel结构调整、读写策略优化、流水线设计等手段,实现了比一开始的设计方案性能提升6倍的FPGA加速器,且该加速器性能在OGB Products和OGB Proteins数据集上达到了和CPU(Xeon(R)Silver 4214)48线程相比×1.22到×1.99倍的加速
C++ INI Text Makefile Python other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》