Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
SoldierChen authored Dec 1, 2020
1 parent 7ab614c commit d074802
Showing 1 changed file with 3 additions and 7 deletions.
10 changes: 3 additions & 7 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
[![GitHub license](https://img.shields.io/badge/license-apache2-yellowgreen)](./LICENSE)
[![GitHub issues](https://img.shields.io/github/issues/Xtra-Computing/On-the-fly-data-shuffling-for-OpenCL-based-FPGAs.svg)](https://github.com/Xtra-Computing/On-the-fly-data-shuffling-for-OpenCL-based-FPGAs/issues)

# ThunderGP: Fast Graph Processing for HLS-based FPGAs
# ThunderGP: HLS-based Graph Processing Framework on FPGAs

## What's new?

Expand All @@ -22,11 +22,7 @@ ThunderGP enables data scientists to enjoy the ***performance*** of FPGA-based g
Two aspacts make the ThunderGP deliver superior performance.
On the one hand, ThunderGP embraces an improved execution flow to better exploit the pipeline parallelism of FPGA and alleviate the data access amount to the global memory. On the other hand, the memory accesses are highly optimized to fully utilize the memory bandwidth capacity of the hardware platforms.

ThunderGP can run on both Xilinx and Intel platforms:

* [Check the implementation on Intel platform out.](https://github.com/Xtra-Computing/On-the-fly-data-shuffling-for-OpenCL-based-FPGAs/)

* On Xilinx multi-SLR based FPGAs, it is running at 250Mhz, and the performance can be up to ***6400 MTEPS (million traversed edges per second)***, or a ***2.9 times speedup*** over the state-of-the-art.
On Xilinx multi-SLR based FPGAs, it is running at 250Mhz, and the performance can be up to ***6400 MTEPS (million traversed edges per second)***, or a ***2.9 times speedup*** over the state-of-the-art.


## Prerequisites
Expand Down Expand Up @@ -59,7 +55,7 @@ Here is an example of implementing PR algorithm.
$ cd ./
$ make app=pr clean
$ make app=pr all # make the host execution program and FPGA execution program for pagerank application. It takes time.
$ ./host [bitfile] [graph name] #e.g., ./host_graph_fpga _x/link/int/graph_fpga.hw.xilinx_vcu1525_xdma_201830_1.xclbin wiki-talk
$ ./host_graph_fpga_pr ./xclbin_pr/graph_fpga.hw.xilinx_vcu1525_xdma_201830_1.xclbin wiki-talk
```
#### More details: [Compiling ThunderGP ](docs/compile_arch.md)

Expand Down

0 comments on commit d074802

Please sign in to comment.