English:

FHE GPU Acceleration Library

This project is based on the framework of Microsoft SEAL library, and the original interface is completely retained

Users do not need to modify the code as you are using the original SEAL, and can directly benifit from this project for GPU acceleration calculation.

Acceleration

More details and test results can be found in our paper.

Preparation

1. Install CUDA Download and install CUDA 11.7 https://developer.nvidia.com/cuda-11-7-0-download-archive

2. Install CMake Install CMake 3.18.3

Compilation

Execute in the root directory

cmake -S . -B build
cd build
make 

Test:

cd build/bin
./sealexamples

中文：

FHE GPU加速库

该项目基于微软SEAL库框架进行完成，完全保留原有接口

用户无需修改原有使用SEAL时的代码，可以直接使用本项目进行GPU加速计算。

加速效果

加速效果和细节在我们的论文中可详细查阅。

准备操作

1. 安装CUDA 下载安装11.7版本CUDA https://developer.nvidia.com/cuda-11-7-0-download-archive
2. 安装CMake 安装3.18.3版本cmake

编译方法

根目录下执行

cmake -S . -B build
cd build
make 

测试：

cd build/bin
./sealexamples