diff --git a/doc/RUN_IN_DOCKER.md b/doc/RUN_IN_DOCKER.md
new file mode 100644
index 0000000000000000000000000000000000000000..345aabed52cb30282057ea7f5ba4953a9681d6d8
--- /dev/null
+++ b/doc/RUN_IN_DOCKER.md
@@ -0,0 +1,175 @@
+# How to run PaddleServing in Docker
+
+## Requirements
+
+Docker (GPU version requires nvidia-docker to be installed on the GPU machine)
+
+## CPU
+
+### Get docker image
+
+You can get images in two ways:
+
+1. Pull image directly
+
+   ```bash
+   docker pull hub.baidubce.com/ctr/paddleserving:0.1.3
+   ```
+
+2. Building image based on dockerfile
+
+   Create a new folder and copy [Dockerfile](../tools/Dockerfile) to this folder, and run the following command:
+
+   ```bash
+   docker build -t hub.baidubce.com/ctr/paddleserving:0.1.3 .
+   ```
+
+### Create container
+
+```bash
+docker run -p 9292:9292 --name test -dit hub.baidubce.com/ctr/paddleserving:0.1.3
+docker exec -it test bash
+```
+
+The `-p` option is to map the `9292` port of the container to the `9292` port of the host.
+
+### Install PaddleServing
+
+In order to make the image smaller, the PaddleServing package is not installed in the image. You can run the following command to install it
+
+```bash
+pip install paddle-serving-server
+```
+
+### Test example
+
+Get the trained Boston house price prediction model by the following command:
+
+```bash
+wget --no-check-certificate https://paddle-serving.bj.bcebos.com/uci_housing.tar.gz
+tar -xzf uci_housing.tar.gz
+```
+
+- Test HTTP service
+
+  Running on the Server side (inside the container):
+
+  ```bash
+  python -m paddle_serving_server.web_serve --model uci_housing_model --thread 10 --port 9292 --name uci &>std.log 2>err.log &
+  ```
+
+  Running on the Client side (inside or outside the container):
+
+  ```bash
+  curl -H "Content-Type:application/json" -X POST -d '{"x": [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727, -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332], "fetch":["price"]}' http://127.0.0.1:9292/uci/prediction
+  ```
+
+- Test RPC service
+
+  Running on the Server side (inside the container):
+
+  ```bash
+  python -m paddle_serving_server.serve --model uci_housing_model --thread 10 --port 9292 &>std.log 2>err.log &
+  ```
+
+  Running following Python code on the Client side (inside or outside the container, The `paddle-serving-client` package needs to be installed):
+
+  ```bash
+  from paddle_serving_client import Client
+  
+  client = Client()
+  client.load_client_config("uci_housing_client/serving_client_conf.prototxt")
+  client.connect(["127.0.0.1:9292"])
+  data = [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727,
+          -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332]
+  fetch_map = client.predict(feed={"x": data}, fetch=["price"])
+  print(fetch_map)
+  ```
+
+  
+
+## GPU
+
+The GPU version is basically the same as the CPU version, with only some differences in interface naming (GPU version requires nvidia-docker to be installed on the GPU machine).
+
+### Get docker image
+
+You can also get images in two ways:
+
+1. Pull image directly
+
+   ```bash
+   nvidia-docker pull hub.baidubce.com/ctr/paddleserving:0.1.3-gpu
+   ```
+
+2. Building image based on dockerfile
+
+   Create a new folder and copy [Dockerfile.gpu](../tools/Dockerfile.gpu) to this folder, and run the following command:
+
+   ```bash
+   nvidia-docker build -t hub.baidubce.com/ctr/paddleserving:0.1.3-gpu .
+   ```
+
+### Create container
+
+```bash
+nvidia-docker run -p 9292:9292 --name test -dit hub.baidubce.com/ctr/paddleserving:0.1.3-gpu
+nvidia-docker exec -it test bash
+```
+
+The `-p` option is to map the `9292` port of the container to the `9292` port of the host.
+
+### Install PaddleServing
+
+In order to make the image smaller, the PaddleServing package is not installed in the image. You can run the following command to install it:
+
+```bash
+pip install paddle-serving-server-gpu
+```
+
+### Test example
+
+Get the trained Boston house price prediction model by the following command:
+
+```bash
+wget --no-check-certificate https://paddle-serving.bj.bcebos.com/uci_housing.tar.gz
+tar -xzf uci_housing.tar.gz
+```
+
+- Test HTTP service
+
+  Running on the Server side (inside the container):
+
+  ```bash
+  python -m paddle_serving_server_gpu.web_serve --model uci_housing_model --thread 10 --port 9292 --name uci
+  ```
+
+  Running on the Client side (inside or outside the container):
+
+  ```bash
+  curl -H "Content-Type:application/json" -X POST -d '{"x": [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727, -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332], "fetch":["price"]}' http://127.0.0.1:9292/uci/prediction
+  ```
+
+- Test RPC service
+
+  Running on the Server side (inside the container):
+
+  ```bash
+  python -m paddle_serving_server_gpu.serve --model uci_housing_model --thread 10 --port 9292
+  ```
+
+  Running following Python code on the Client side (inside or outside the container, The `paddle-serving-client` package needs to be installed):
+
+  ```bash
+  from paddle_serving_client import Client
+  
+  client = Client()
+  client.load_client_config("uci_housing_client/serving_client_conf.prototxt")
+  client.connect(["127.0.0.1:9292"])
+  data = [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727,
+          -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332]
+  fetch_map = client.predict(feed={"x": data}, fetch=["price"])
+  print(fetch_map)
+  ```
+
+  
diff --git a/doc/RUN_IN_DOCKER_CN.md b/doc/RUN_IN_DOCKER_CN.md
new file mode 100644
index 0000000000000000000000000000000000000000..7e2f28bdbd73c793faee96def6b625e1bbff2ba9
--- /dev/null
+++ b/doc/RUN_IN_DOCKER_CN.md
@@ -0,0 +1,171 @@
+# 如何在Docker中运行PaddleServing
+
+## 环境要求
+
+Docker（GPU版本需要在GPU机器上安装nvidia-docker）
+
+## CPU版本
+
+### 获取镜像
+
+可以通过两种方式获取镜像。
+
+1. 直接拉取镜像
+
+   ```bash
+   docker pull hub.baidubce.com/ctr/paddleserving:0.1.3
+   ```
+
+2. 基于Dockerfile构建镜像
+
+   建立新目录，复制[Dockerfile](../tools/Dockerfile)内容到该目录下Dockerfile文件。执行
+
+   ```bash
+   docker build -t hub.baidubce.com/ctr/paddleserving:0.1.3 .
+   ```
+
+### 创建容器并进入
+
+```bash
+docker run -p 9292:9292 --name test -dit hub.baidubce.com/ctr/paddleserving:0.1.3
+docker exec -it test bash
+```
+
+`-p`选项是为了将容器的`9292`端口映射到宿主机的`9292`端口。
+
+### 安装PaddleServing
+
+为了减小镜像的体积，镜像中没有安装Serving包，要执行下面命令进行安装
+
+```bash
+pip install paddle-serving-server
+```
+
+### 测试example
+
+通过下面命令获取训练好的Boston房价预估模型：
+
+```bash
+wget --no-check-certificate https://paddle-serving.bj.bcebos.com/uci_housing.tar.gz
+tar -xzf uci_housing.tar.gz
+```
+
+- 测试HTTP服务
+
+  在Server端（容器内）运行：
+
+  ```bash
+  python -m paddle_serving_server.web_serve --model uci_housing_model --thread 10 --port 9292 --name uci &>std.log 2>err.log &
+  ```
+
+  在Client端（容器内或容器外）运行：
+
+  ```bash
+  curl -H "Content-Type:application/json" -X POST -d '{"x": [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727, -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332], "fetch":["price"]}' http://127.0.0.1:9292/uci/prediction
+  ```
+
+- 测试RPC服务
+
+  在Server端（容器内）运行：
+
+  ```bash
+  python -m paddle_serving_server.serve --model uci_housing_model --thread 10 --port 9292 &>std.log 2>err.log &
+  ```
+
+  在Client端（容器内或容器外，需要安装`paddle-serving-client`包）运行下面Python代码：
+
+  ```python
+  from paddle_serving_client import Client
+  
+  client = Client()
+  client.load_client_config("uci_housing_client/serving_client_conf.prototxt")
+  client.connect(["127.0.0.1:9292"])
+  data = [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727,
+          -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332]
+  fetch_map = client.predict(feed={"x": data}, fetch=["price"])
+  print(fetch_map)
+  ```
+
+## GPU版本
+
+GPU版本与CPU版本基本一致，只有部分接口命名的差别（GPU版本需要在GPU机器上安装nvidia-docker）。
+
+### 获取镜像
+
+可以通过两种方式获取镜像。
+
+1. 直接拉取镜像
+
+   ```bash
+   nvidia-docker pull hub.baidubce.com/ctr/paddleserving:0.1.3-gpu
+   ```
+
+2. 基于Dockerfile构建镜像
+
+   建立新目录，复制[Dockerfile.gpu](../tools/Dockerfile.gpu)内容到该目录下Dockerfile文件。执行
+
+   ```bash
+   nvidia-docker build -t hub.baidubce.com/ctr/paddleserving:0.1.3-gpu .
+   ```
+
+### 创建容器并进入
+
+```bash
+nvidia-docker run -p 9292:9292 --name test -dit hub.baidubce.com/ctr/paddleserving:0.1.3-gpu
+nvidia-docker exec -it test bash
+```
+
+`-p`选项是为了将容器的`9292`端口映射到宿主机的`9292`端口。
+
+### 安装PaddleServing
+
+为了减小镜像的体积，镜像中没有安装Serving包，要执行下面命令进行安装
+
+```bash
+pip install paddle-serving-server-gpu
+```
+
+### 测试example
+
+通过下面命令获取训练好的Boston房价预估模型：
+
+```bash
+wget --no-check-certificate https://paddle-serving.bj.bcebos.com/uci_housing.tar.gz
+tar -xzf uci_housing.tar.gz
+```
+
+- 测试HTTP服务
+
+  在Server端（容器内）运行：
+
+  ```bash
+  python -m paddle_serving_server_gpu.web_serve --model uci_housing_model --thread 10 --port 9292 --name uci 
+  ```
+
+  在Client端（容器内或容器外）运行：
+
+  ```bash
+  curl -H "Content-Type:application/json" -X POST -d '{"x": [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727, -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332], "fetch":["price"]}' http://127.0.0.1:9292/uci/prediction
+  ```
+
+- 测试RPC服务
+
+  在Server端（容器内）运行：
+
+  ```bash
+  python -m paddle_serving_server_gpu.serve --model uci_housing_model --thread 10 --port 9292
+  ```
+
+  在Client端（容器内或容器外，需要安装`paddle-serving-client`包）运行下面Python代码：
+
+  ```bash
+  from paddle_serving_client import Client
+  
+  client = Client()
+  client.load_client_config("uci_housing_client/serving_client_conf.prototxt")
+  client.connect(["127.0.0.1:9292"])
+  data = [0.0137, -0.1136, 0.2553, -0.0692, 0.0582, -0.0727,
+          -0.1583, -0.0584, 0.6283, 0.4919, 0.1856, 0.0795, -0.0332]
+  fetch_map = client.predict(feed={"x": data}, fetch=["price"])
+  print(fetch_map)
+  ```
diff --git a/tools/Dockerfile b/tools/Dockerfile
index 359dd52c0726e9a421138bf3ecf4d6cff3b2036f..a39ce5bb76e411edeb94766d0c9aae23c6e7e62f 100644
--- a/tools/Dockerfile
+++ b/tools/Dockerfile
@@ -1,24 +1,8 @@
-FROM centos:centos6.10
-RUN yum -y install wget \
-    && wget http://people.centos.org/tru/devtools-2/devtools-2.repo -O /etc/yum.repos.d/devtoolset-2.repo \
-    && yum -y install devtoolset-2-gcc devtoolset-2-gcc-c++ devtoolset-2-binutils \
-    && source /opt/rh/devtoolset-2/enable \
-    && echo "source /opt/rh/devtoolset-2/enable" >> /etc/profile \
-    && yum -y install git openssl-devel curl-devel bzip2-devel \
-    && wget https://cmake.org/files/v3.5/cmake-3.5.2.tar.gz \
-    && tar xvf cmake-3.5.2.tar.gz \
-    && cd cmake-3.5.2 \
-    &&  ./bootstrap --prefix=/usr \
-    && make \
-    && make install \
-    && cd .. \
-    && rm -r cmake-3.5.2* \
-    && wget https://dl.google.com/go/go1.12.12.linux-amd64.tar.gz \
-    && tar -xzvf go1.12.12.linux-amd64.tar.gz \
-    && mv go /usr/local/go \
-    && rm go1.12.12.linux-amd64.tar.gz \
-    && echo "export GOROOT=/usr/local/go" >> /root/.bashrc \
-    && echo "export GOPATH=$HOME/go" >> /root/.bashrc \
-    && echo "export PATH=$PATH:/usr/local/go/bin" >> /root/.bashrc
-
+FROM centos:7.3.1611
 
+RUN yum -y install wget && \
+    yum -y install epel-release && yum -y install patchelf && \
+    yum -y install gcc make python-devel && \
+    yum clean all && \
+    curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py && \
+    python get-pip.py && rm get-pip.py
diff --git a/tools/Dockerfile.devel b/tools/Dockerfile.devel
new file mode 100644
index 0000000000000000000000000000000000000000..a4b5b5fe48b5c4d5c74d66dc688fa5d594a33266
--- /dev/null
+++ b/tools/Dockerfile.devel
@@ -0,0 +1,22 @@
+FROM centos:7.3.1611
+RUN yum -y install wget >/dev/null \
+    && yum -y install gcc gcc-c++ make glibc-static which >/dev/null \
+    && yum -y install git openssl-devel curl-devel bzip2-devel python-devel >/dev/null \
+    && wget https://cmake.org/files/v3.2/cmake-3.2.0-Linux-x86_64.tar.gz >/dev/null \
+    && tar xzf cmake-3.2.0-Linux-x86_64.tar.gz \
+    && mv cmake-3.2.0-Linux-x86_64 /usr/local/cmake3.2.0 \
+    && echo 'export PATH=/usr/local/cmake3.2.0/bin:$PATH' >> /root/.bashrc \
+    && rm cmake-3.2.0-Linux-x86_64.tar.gz \
+    && wget https://dl.google.com/go/go1.14.linux-amd64.tar.gz >/dev/null \
+    && tar xzf go1.14.linux-amd64.tar.gz \
+    && mv go /usr/local/go \
+    && echo 'export GOROOT=/usr/local/go' >> /root/.bashrc \
+    && echo 'export PATH=/usr/local/go/bin:$PATH' >> /root/.bashrc \
+    && rm go1.14.linux-amd64.tar.gz \
+    && yum -y install python-devel sqlite-devel >/dev/null \
+    && curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py >/dev/null \
+    && python get-pip.py >/dev/null \
+    && pip install google protobuf setuptools wheel flask >/dev/null \
+    && rm get-pip.py \
+    && yum -y install epel-release && yum -y install patchelf \
+    && yum clean all
diff --git a/tools/Dockerfile.gpu b/tools/Dockerfile.gpu
index 5258854691866ab9357a947f42081ae8db6984d7..427ae83bcb805ec70c1e6d575e84234f17e9fb30 100644
--- a/tools/Dockerfile.gpu
+++ b/tools/Dockerfile.gpu
@@ -1,11 +1,15 @@
-FROM paddlepaddle/paddle_manylinux_devel:cuda9.0_cudnn7
-RUN yum -y install git openssl-devel curl-devel bzip2-devel \
-    && wget https://dl.google.com/go/go1.12.12.linux-amd64.tar.gz \
-    && tar -xzvf go1.12.12.linux-amd64.tar.gz \
-    && rm -rf /usr/local/go \
-    && mv go /usr/local/go \
-    && rm go1.12.12.linux-amd64.tar.gz \
-    && echo "GOROOT=/usr/local/go" >> /root/.bashrc \
-    && echo "GOPATH=$HOME/go" >> /root/.bashrc \
-    && echo "PATH=$PATH:$GOROOT/bin" >> /root/.bashrc
+FROM nvidia/cuda:9.0-cudnn7-runtime-centos7
 
+RUN yum -y install wget && \
+    yum -y install epel-release && yum -y install patchelf && \
+    yum -y install gcc make python-devel && \
+    yum -y install libSM-1.2.2-2.el7.x86_64 --setopt=protected_multilib=false && \
+    yum -y install libXrender-0.9.10-1.el7.x86_64 --setopt=protected_multilib=false && \
+    yum -y install libXext-1.3.3-3.el7.x86_64 --setopt=protected_multilib=false && \
+    yum clean all && \
+    curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py && \
+    python get-pip.py && rm get-pip.py && \
+    ln -s /usr/local/cuda-9.0/lib64/libcublas.so.9.0 /usr/local/cuda-9.0/lib64/libcublas.so && \
+    echo 'export LD_LIBRARY_PATH=/usr/local/cuda/lib64':$LD_LIBRARY_PATH >> /root/.bashrc && \
+    ln -s /usr/local/cuda-9.0/targets/x86_64-linux/lib/libcudnn.so.7 /usr/local/cuda-9.0/targets/x86_64-linux/lib/libcudnn.so && \
+    echo 'export LD_LIBRARY_PATH=/usr/local/cuda-9.0/targets/x86_64-linux/lib:$LD_LIBRARY_PATH' >> /root/.bashrc