Files
FastDeploy/serving/docs/zh_CN/compile.md
T
heliqi e79dc863e4 [Doc][Serving]serving doc update version to 1.0.0 (#755)
serving doc update version to 1.0.0
2022-11-30 14:41:42 +08:00

45 lines
1.5 KiB
Markdown

# 服务化部署镜像编译
本文档介绍如何制作FastDploy镜像
## 制作GPU镜像
FastDploy发布的GPU镜像基于[Triton Inference Server](https://github.com/triton-inference-server/server)的21.10版本进行制作,如果有其他CUDA版本需求,可以参照[NVIDIA 官网](https://docs.nvidia.com/deeplearning/frameworks/support-matrix/index.html)中展示的版本信息修改Dockerfile和scripts中的脚本.
```
# 进入serving目录执行脚本编译fastdeploy和服务化的backend
cd serving
bash scripts/build.sh
# 退出到FastDeploy主目录,制作镜像
# x.y.z为FastDeploy版本号,可根据情况自己确定。比如: 1.0.0
cd ../
docker build -t paddlepaddle/fastdeploy:x.y.z-gpu-cuda11.4-trt8.4-21.10 -f serving/Dockerfile .
```
## 制作CPU镜像
```
# 进入serving目录执行脚本编译fastdeploy和服务化的backend
cd serving
bash scripts/build.sh OFF
# 退出到FastDeploy主目录,制作镜像
# x.y.z为FastDeploy版本号,可根据情况自己确定。比如: 1.0.0
cd ../
docker build -t paddlepaddle/fastdeploy:x.y.z-cpu-only-21.10 -f serving/Dockerfile_cpu .
```
## 制作IPU镜像
```
# 进入serving目录执行脚本编译fastdeploy和服务化的backend
cd serving
bash scripts/build_fd_ipu.sh
# 退出到FastDeploy主目录,制作镜像
# x.y.z为FastDeploy版本号,可根据情况自己确定。比如: 1.0.0
cd ../
docker build -t paddlepaddle/fastdeploy:x.y.z-ipu-only-21.10 -f serving/Dockerfile_ipu .
```