change: upgrade MMS version and update command (#71)

Co-authored-by: akulk314 <[email protected]>
Co-authored-by: Ajay Karpur <[email protected]>

Author: 3 people

README.md

@@ -18,7 +18,7 @@ A container provides an effectively isolated environment, ensuring a consistent
 Containerizing your model and code enables fast and reliable deployment of your model.
 
 The **SageMaker Inference Toolkit** implements a model serving stack and can be easily added to any Docker container, making it [deployable to SageMaker](https://aws.amazon.com/sagemaker/deploy/).
-This library's serving stack is built on [Multi Model Server](https://github.com/awslabs/mxnet-model-server), and it can serve your own models or those you trained on SageMaker using [machine learning frameworks with native SageMaker support](https://docs.aws.amazon.com/sagemaker/latest/dg/frameworks.html).
+This library's serving stack is built on [Multi Model Server](https://github.com/awslabs/multi-model-server), and it can serve your own models or those you trained on SageMaker using [machine learning frameworks with native SageMaker support](https://docs.aws.amazon.com/sagemaker/latest/dg/frameworks.html).
 If you use a [prebuilt SageMaker Docker image for inference](https://docs.aws.amazon.com/sagemaker/latest/dg/pre-built-containers-frameworks-deep-learning.html), this library may already be included.
 
 For more information, see the Amazon SageMaker Developer Guide sections on [building your own container with Multi Model Server](https://docs.aws.amazon.com/sagemaker/latest/dg/build-multi-model-build-container.html) and [using your own models](https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms.html).
@@ -98,7 +98,7 @@ To use the SageMaker Inference Toolkit, you need to do the following:
 
 2.  Implement a handler service that is executed by the model server.
     ([Here is an example](https://github.com/aws/sagemaker-pytorch-serving-container/blob/master/src/sagemaker_pytorch_serving_container/handler_service.py) of a handler service.)
-    For more information on how to define your `HANDLER_SERVICE` file, see [the MMS custom service documentation](https://github.com/awslabs/mxnet-model-server/blob/master/docs/custom_service.md).
+    For more information on how to define your `HANDLER_SERVICE` file, see [the MMS custom service documentation](https://github.com/awslabs/multi-model-server/blob/master/docs/custom_service.md).
 
     ``` python
     from sagemaker_inference.default_handler_service import DefaultHandlerService
@@ -112,7 +112,7 @@ To use the SageMaker Inference Toolkit, you need to do the following:
         This class extends ``DefaultHandlerService``, which define the following:
             - The ``handle`` method is invoked for all incoming inference requests to the model server.
             - The ``initialize`` method is invoked at model server start up.
-        Based on: https://github.com/awslabs/mxnet-model-server/blob/master/docs/custom_service.md
+        Based on: https://github.com/awslabs/multi-model-server/blob/master/docs/custom_service.md
         """
         def __init__(self):
             transformer = Transformer(default_inference_handler=DefaultPytorchInferenceHandler())

src/sagemaker_inference/default_handler_service.py

@@ -27,7 +27,7 @@ class DefaultHandlerService(object):
         - The ``handle`` method is invoked for all incoming inference requests to the model server.
         - The ``initialize`` method is invoked at model server start up.
 
-    Implementation of: https://github.com/awslabs/mxnet-model-server/blob/master/docs/custom_service.md
+    Implementation of: https://github.com/awslabs/multi-model-server/blob/master/docs/custom_service.md
     """
 
     def __init__(self, transformer=None):

src/sagemaker_inference/etc/default-mms.properties

@@ -1,4 +1,4 @@
-# Based on: https://github.com/awslabs/mxnet-model-server/blob/master/docs/configuration.md
+# Based on: https://github.com/awslabs/multi-model-server/blob/master/docs/configuration.md
 enable_envvars_config=true
 decode_input_request=false
 load_models=ALL

src/sagemaker_inference/model_server.py

@@ -79,8 +79,8 @@ def start_model_server(handler_service=DEFAULT_HANDLER_SERVICE):
     if os.path.exists(REQUIREMENTS_PATH):
         _install_requirements()
 
-    mxnet_model_server_cmd = [
-        "mxnet-model-server",
+    multi_model_server_cmd = [
+        "multi-model-server",
         "--start",
         "--model-store",
         MODEL_STORE,
@@ -90,8 +90,8 @@ def start_model_server(handler_service=DEFAULT_HANDLER_SERVICE):
         DEFAULT_MMS_LOG_FILE,
     ]
 
-    logger.info(mxnet_model_server_cmd)
-    subprocess.Popen(mxnet_model_server_cmd)
+    logger.info(multi_model_server_cmd)
+    subprocess.Popen(multi_model_server_cmd)
 
     mms_process = _retrieve_mms_server_process()
 

test/container/dummy/Dockerfile

@@ -11,7 +11,7 @@ ENV SAGEMAKER_MULTI_MODEL=true
 ENV SAGEMAKER_BIND_TO_PORT=${SAGEMAKER_BIND_TO_PORT:-8080}
 
 # Update MMS version
-RUN pip3 install mxnet-model-server==1.0.8
+RUN pip3 install multi-model-server
 
 # Install Mxnet (for handler_service)
 RUN pip3 install mxnet

test/container/mxnet/Dockerfile

@@ -9,7 +9,7 @@ LABEL com.amazonaws.sagemaker.capabilities.accept-bind-to-port=true
 # https://docs.aws.amazon.com/sagemaker/latest/dg/build-multi-model-build-container.html
 LABEL com.amazonaws.sagemaker.capabilities.multi-models=true
 
-ARG MMS_VERSION=1.0.8
+ARG MMS_VERSION=1.1.2
 ARG MX_URL=https://aws-mxnet-pypi.s3-us-west-2.amazonaws.com/1.6.0/aws_mxnet_mkl-1.6.0-py2.py3-none-manylinux1_x86_64.whl
 ARG PYTHON=python3
 ARG PYTHON_PIP=python3-pip
@@ -73,7 +73,7 @@ COPY mxnet/sagemaker_inference.tar.gz /sagemaker_inference.tar.gz
 RUN ${PIP} install --no-cache-dir \
     ${MX_URL} \
     git+git://github.com/dmlc/[email protected] \
-    mxnet-model-server==$MMS_VERSION \
+    multi-model-server==$MMS_VERSION \
     keras-mxnet==2.2.4.1 \
     numpy==1.17.4 \
     onnx==1.4.1 \
@@ -122,4 +122,4 @@ RUN curl https://aws-dlc-licenses.s3.amazonaws.com/aws-mxnet-1.6.0/license.txt -
 
 EXPOSE 8080 8081
 ENTRYPOINT ["python", "/usr/local/bin/dockerd_entrypoint.py"]
-CMD ["mxnet-model-server", "--start", "--mms-config", "/home/model-server/config.properties"]
+CMD ["multi-model-server", "--start", "--mms-config", "/home/model-server/config.properties"]

test/unit/test_model_server.py

@@ -50,8 +50,8 @@ def test_start_model_server_default_service_handler(
     exists.assert_called_once_with(REQUIREMENTS_PATH)
     install_requirements.assert_called_once_with()
 
-    mxnet_model_server_cmd = [
-        "mxnet-model-server",
+    multi_model_server_cmd = [
+        "multi-model-server",
         "--start",
         "--model-store",
         model_server.DEFAULT_MMS_MODEL_DIRECTORY,
@@ -61,7 +61,7 @@ def test_start_model_server_default_service_handler(
         model_server.DEFAULT_MMS_LOG_FILE,
     ]
 
-    subprocess_popen.assert_called_once_with(mxnet_model_server_cmd)
+    subprocess_popen.assert_called_once_with(multi_model_server_cmd)
     sigterm.assert_called_once_with(retrieve.return_value)