Chore: Reorg dockerfiles and fix docs (#51)

.github/workflows/push_docker.yaml

@@ -14,7 +14,7 @@ on:
       - 'zeus/**'
       - 'zeus_monitor/**'
       - '.dockerignore'
-      - 'Dockerfile'
+      - 'docker/Dockerfile'
       - 'LICENSE'
       - 'setup.py'
       - 'pyproject.toml'
@@ -52,7 +52,7 @@ jobs:
         uses: docker/build-push-action@v3
         with:
           context: .
-          file: Dockerfile
+          file: docker/Dockerfile
           builder: ${{ steps.buildx.outputs.name }}
           push: true
           tags: ${{ steps.meta.outputs.tags }}

docker/docker-compose.yaml → docker/batch_size_optimizer/server-docker-compose.yaml

@@ -5,8 +5,8 @@ services:
   server: 
     image: bso-server
     build:
-      context: ../
-      dockerfile: ./docker/bso_server.Dockerfile
+      context: ../../
+      dockerfile: ./docker/batch_size_optimizer/server.Dockerfile
     container_name: bso
     restart: always
     environment:
@@ -55,8 +55,8 @@ services:
   migration:
     image: bso-migration
     build:
-      context: ../
-      dockerfile: ./docker/bso_migration.Dockerfile
+      context: ../../
+      dockerfile: ./docker/batch_size_optimizer/migration.Dockerfile
     deploy:
       restart_policy:
         condition: on-failure

docs/batch_size_optimizer/index.md

@@ -74,8 +74,8 @@ sequenceDiagram;
 
             ```Shell
             # From the root directory
-            docker build -f ./docker/bso_server.Dockerfile -t bso-server . 
-            docker build -f ./docker/bso_migration.Dockerfile -t bso-migration .
+            docker build -f ./docker/batch_size_optimizer/server.Dockerfile -t bso-server . 
+            docker build -f ./docker/batch_size_optimizer/migration.Dockerfile -t bso-migration .
             ```
 
         2. Create Kubernetes yaml files using Kompose. Kompose is a tool that converts docker-compose files into Kubernetes files. For more information, visit [Kompose Reference](#kompose-references)
@@ -146,7 +146,7 @@ sequenceDiagram;
 ### Remark about the server
 
 Zeus Batch Size Optimizer server is using Sqlalchemy to support various types of databases. However, you need to download the corresponding async connection driver.
-As a default, we are using Mysql. You can add installation code to `bso_migration.Dockerfile` and `bso_server.Dockerfile`. Refer to those files for reference.
+As a default, we are using Mysql. You can add installation code to `docker/batch_size_optimizer/migration.Dockerfile` and `docker/batch_size_optimizer/server.Dockerfile`. Refer to those files for reference.
 
 ## Use BSO in your training script (Client)
 

examples/bso_server/README.md → examples/batch_size_optimizer/README.md

@@ -5,7 +5,7 @@ Batch size optimzer is composed of two parts: server and client. Client will be
 ## Data parallel training with Zeus
 
 In the case of data parallel training, Batch size optimizer should be able to give the consistent batch size to all gpus. Since there is no way for batch size to tell the differences between concurrent job submissions and multiple GPU training, we ask users to send a request from a single GPU and broadcast the result(batch size, trial number) to other GPUs. In the case of reporting the result to the batch size optimizer server and receiving the corresponding result (train fail or succeeded) can be dealt by the server since it has the `trial_number`. Thus, report doesn't require any broadcast or communications with other GPUs.
-Refer to the `examples/bso_server/mnist_dp.py` for the use case.
+Refer to the `examples/batch_size_optimizer/mnist_dp.py` for the use case.
 
 ## Kubeflow
 
@@ -19,7 +19,7 @@ Kubeflow is a tool to easily deploy your ML workflows to kubernetes. We provides
 
     ```Shell
     # From project root directory
-    docker build -f ./examples/bso_server/mnist.Dockerfile -t mnist-example . 
+    docker build -f ./examples/batch_size_optimizer/mnist.Dockerfile -t mnist-example . 
     ```
 
 3. Deploy training script.

examples/bso_server/mnist_dp.yaml → examples/batch_size_optimizer/mnist_dp.yaml

@@ -16,7 +16,7 @@ spec:
               imagePullPolicy: Never
               command:
                 - "python3"
-                - "/workspace/examples/bso_server/mnist_dp.py"
+                - "/workspace/examples/batch_size_optimizer/mnist_dp.py"
                 - "--epochs=5"
                 - "--backend=nccl"
               env:
@@ -38,7 +38,7 @@ spec:
               imagePullPolicy: Never
               command:
                 - "python3"
-                - "/workspace/examples/bso_server/mnist_dp.py"
+                - "/workspace/examples/batch_size_optimizer/mnist_dp.py"
                 - "--epochs=5"
                 - "--backend=nccl"
               env:
@@ -48,4 +48,4 @@ spec:
                   value: "mnist-dev-dp-2"
               securityContext:
                 capabilities:
-                  add: ["SYS_ADMIN"]
+                  add: ["SYS_ADMIN"]

examples/bso_server/mnist_single_gpu.yaml → examples/batch_size_optimizer/mnist_single_gpu.yaml

@@ -16,7 +16,7 @@ spec:
               imagePullPolicy: Never
               command:
                 - "python3"
-                - "/workspace/examples/bso_server/mnist_single_gpu.py"
+                - "/workspace/examples/batch_size_optimizer/mnist_single_gpu.py"
                 - "--epochs=5"
               env:
                 - name: ZEUS_SERVER_URL
@@ -27,4 +27,4 @@ spec:
                   value: "INFO"
               securityContext:
                 capabilities:
-                  add: ["SYS_ADMIN"]
+                  add: ["SYS_ADMIN"]