Adlik

Adlik [ædlik] is an end-to-end optimizing framework for deep learning models. The goal of Adlik is to accelerate deep learning inference process both on cloud and embedded environment.

With Adlik framework, different deep learning models can be deployed to different platforms with high performance in a much flexible and easy way.

In cloud environment, the compiled model and Adlik Inference Engine should be built as a docker image, and deployed as a container.
In edge environment, Adlik Inference Engine should be deployed as a container. The compiled model should be transferred to edge environment, and the Adlik Inference Engine should automatically update and load model.
In device environment, Adlik Inference Engine and the compiled model should be compiled into a binary file (so or lib). Users who want to run model inference on device should link user defined AI function and Adlik binary file to the execution file, and run directly.

Inference performance of Adlik

We test the inference performance of Adlik on the same CPU or GPU using the simple CNN model (MNIST model), the ResNet50 model, and InceptionV3 with different serving engines. The test performance data of Adlik on different models are as follows:

Model Optimizer

Model optimizer focuses on specific hardware and runs on it to achieve acceleration. The proposed framework mainly consists of two categories of algorithm components, i.e. pruner and quantizer.

Model Compiler

Model compiler supports several optimizing technologies like pruning, quantization and structural compression, which can be easily used for models developed with TensorFlow, Keras, PyTorch, etc.

Serving Engine

Serving Engine provides deep learning models with optimized runtime based on the deployment environment. Put simply, based on a deep learning model, the users of Adlik can optimize it with model compiler and then deploy it to a certain platform with Adlik serving platform.

Getting Started

Docker images

All Adlik compiler images and serving images are stored in Alibaba Cloud. These images can be downloaded and used directly, users do not need to build the Adlik on Ubuntu. Users can use the compiler images to compile model from H5, CheckPoint, FrozenGraph, ONNX and SavedModel to Openvino, TensorFlow, TensorFlow Lite, TensorRT. Users also can use the serving images for model inference.

Docker pull command:

docker pull docker_image_name:tag

Compiler docker images

The compiler docker images can be used in CPU and GPU. In the CPU, you can compile the model from source type to TensorFlow model, OpenVino model and TensorFlow Lite model. And in the CPU, you can compile the model from source type to TensorFlow model, and TensorRT model. The names and labels of compiler mirrors are as follows, and the first half of label represents the version of TensorRT, the latter part of label represents the version of CUDA:

registry.cn-beijing.aliyuncs.com/adlik/model-compiler:v0.3.0_trt7.2.1.6_cuda11.0

registry.cn-beijing.aliyuncs.com/adlik/model-compiler:v0.3.0_trt7.2.1.6_cuda10.2

Using model compiler image compile model

Run the image.

docker run -it --rm -v source_model:/mnt/model
registry.cn-beijing.aliyuncs.com/adlik/model-compiler:v0.3.0_trt7.2.1.6_cuda11.0 bash

Configure the json file or environment variables required to compile the model.

The config_schema.json describle the json file field information, and for the example, you can reference compiler_json_example.json. For the environment variable field description, see env_field.txt, for the example, reference compiler_env_example.txt.

Note: The checkpoint model must be given the input and output op names of the model when compiling, and other models can be compiled without the input and output op names of the model.

Compile the model.

Compilation instructions (json file mode):

python3 "-c" "import json; import model_compiler as compiler; file=open('/mnt/model/serving_model.json','r');
request = json.load(file);compiler.compile_model(request); file.close()"

Compilation instructions (environment variable mode):

python3 "-c" "import model_compiler.compiler as compiler;compiler.compile_from_env()"

Serving docker images

The serving docker images contains CPU and GPU mirrors. The label of openvino image represents the version of OpenVINO. And for the TensorRT image the first half of label represents the version of TensorRT, the latter part of label represents the version of CUDA. The names and labels of serving mirrors are as follows:

CPU:

registry.cn-beijing.aliyuncs.com/adlik/serving-tflite-cpu:v0.3.0

registry.cn-beijing.aliyuncs.com/adlik/serving-tensorflow-cpu:v0.3.0

registry.cn-beijing.aliyuncs.com/adlik/serving-openvino:v0.3.0

registry.cn-beijing.aliyuncs.com/adlik/serving-tvm-cpu:v0.3.0

GPU:

registry.cn-beijing.aliyuncs.com/adlik/serving-tensorflow-gpu:v0.3.0

registry.cn-beijing.aliyuncs.com/adlik/serving-tftrt-gpu:v0.3.0

registry.cn-beijing.aliyuncs.com/adlik/serving-tensorrt:v0.3.0_trt7.2.1.6_cuda11.0

registry.cn-beijing.aliyuncs.com/adlik/serving-tensorrt:v0.3.0_trt7.2.1.6_cuda10.2

Using the serving images for model inference

Run the mirror and pay attention to mapping out the service port.

docker run -it --rm -p 8500:8500 -v compiled_model:/model
registry.cn-beijing.aliyuncs.com/adlik/serving-openvino:v0.3.0 bash

Load the compiled model in the image and start the service.

adlik-serving --grpc_port=8500 --http_port=8501 --model_base_path=/model

Install the client wheel package adlik serving package or adlik serving gpu package locally, execute the inference code, and perform inference.

Note: If the service port is not mapped when you run the mirror, you need install the adlik serving package or adlik serving gpu package in the container. Then execute the inference code, and perform inference in the container.

Build

This guide is for building Adlik on Ubuntu systems.

First, install Git and Bazel.

Then, clone Adlik and change the working directory into the source directory:

git clone https://github.com/ZTE/Adlik.git
cd Adlik

Build clients

Install the following packages:
- python3-setuptools
- python3-wheel

Build clients:

bazel build //adlik_serving/clients/python:build_pip_package -c opt

Build pip package:

mkdir /tmp/pip-packages && bazel-bin/adlik_serving/clients/python/build_pip_package /tmp/pip-packages

Build serving

First, install the following packages:

automake
libtbb2
libtool
make
python3-six

Build serving with OpenVINO runtime

Install intel-openvino-runtime-ubuntu<OS_VERSION>-<VERSION> package from OpenVINO.

Assume the installation path of OpenVINO is /opt/intel/openvino_VERSION, run the following command:

export INTEL_CVSDK_DIR=/opt/intel/openvino_VERSION
export InferenceEngine_DIR=$INTEL_CVSDK_DIR/deployment_tools/inference_engine/share
bazel build //adlik_serving \
    --config=openvino \
    -c opt

Build serving with TensorFlow CPU runtime

Run the following command:

bazel build //adlik_serving \
    --config=tensorflow-cpu \
    -c opt

Build serving with TensorFlow GPU runtime

Assume building with CUDA version 11.0.

Install the following packages from here and here:
- cuda-cupti-dev-11-0
- libcublas-dev-11-0
- libcudnn8=*+cuda11.0
- libcudnn8-dev=*+cuda11.0
- libcufft-dev-11-0
- libcurand-dev-11-0
- libcusolver-dev-11-0
- libcusparse-dev-11-0
- libnvinfer7=7.2.*+cuda11.0
- libnvinfer-dev=7.2.*+cuda11.0
- libnvinfer-plugin7=7.2.*+cuda11.0
- libnvinfer-plugin-dev=7.2.*+cuda11.0

Run the following command:

env TF_CUDA_VERSION=11.0 TF_NEED_TENSORRT=1 \
    bazel build //adlik_serving \
        --config=tensorflow-gpu \
        -c opt \
        --incompatible_use_specific_tool_files=false

Build serving with TensorFlow Lite CPU runtime

Run the following command:

bazel build //adlik_serving \
    --config=tensorflow-lite-cpu \
    -c opt

Build serving with TensorRT runtime

Assume building with CUDA version 11.0.

Install the following packages from here and here:
- cuda-cupti-dev-11-0
- cuda-nvml-dev-11-0
- cuda-nvrtc-11-0
- libcublas-dev-11-0
- libcudnn8=*+cuda11.0
- libcudnn8-dev=*+cuda11.0
- libcufft-dev-11-0
- libcurand-dev-11-0
- libcusolver-dev-11-0
- libcusparse-dev-11-0
- libnvinfer7=7.2.*+cuda11.0
- libnvinfer-dev=7.2.*+cuda11.0
- libnvonnxparsers7=7.2.*+cuda11.0
- libnvonnxparsers-dev=7.2.*+cuda11.0

Run the following command:

env TF_CUDA_VERSION=11.0 \
    bazel build //adlik_serving \
        --config=TensorRT \
        -c opt \
        --action_env=LIBRARY_PATH=/usr/local/cuda-11.0/lib64/stubs \
        --incompatible_use_specific_tool_files=false

Build serving with TF-TRT runtime

Assume building with CUDA version 11.0.

Install the following packages from here and here:
- cuda-cupti-dev-11-0
- libcublas-dev-11-0
- libcudnn8=*+cuda11.0
- libcudnn8-dev=*+cuda11.0
- libcufft-dev-11-0
- libcurand-dev-11-0
- libcusolver-dev-11-0
- libcusparse-dev-11-0
- libnvinfer7=7.2.*+cuda11.0
- libnvinfer-dev=7.2.*+cuda11.0
- libnvinfer-plugin7=7.2.*+cuda11.0
- libnvinfer-plugin-dev=7.2.*+cuda11.0

Run the following command:

env TF_CUDA_VERSION=11.0 TF_NEED_TENSORRT=1 \
    bazel build //adlik_serving \
        --config=tensorflow-tensorrt \
        -c opt \
        --incompatible_use_specific_tool_files=false

Build serving with Tvm runtime

Install the following packages:
- build-essential
- cmake
- tvm

Run the following command:

bazel build //adlik_serving \
   --config=tvm \
   -c opt

Build in Docker

The ci/docker/build.sh file can be used to build a Docker images that contains all the requirements for building Adlik. You can build Adlik with the Docker image.

Note: If you build the runtime with GPU in a Docker image, you need to add the CUDA environment variables in the Dockerfile, such as:
ENV NVIDIA_VISIBLE_DEVICES all
ENV NVIDIA_DRIVER_CAPABILITIES compute, utility

Release

The version of the service engine Adlik supports.

	TensorFlow 1.14	TensorFlow 2.x	OpenVINO 2021	TensorRT 6	TensorRT 7
Keras	✓	✓	✓	✓	✓
TensorFlow	✓	✓	✓	✓	✓
PyTorch	✗	✗	✓	✓	✓
PaddlePaddle	✓	✓	✓	✓	✓

License

Apache License 2.0

JinXiaozhao/Adlik

Adlik

Inference performance of Adlik

Contents

Model Optimizer

Model Compiler

Serving Engine

Getting Started

Docker images

Compiler docker images

Using model compiler image compile model

Serving docker images

Using the serving images for model inference

Build

Build clients

Build serving

Build serving with OpenVINO runtime

Build serving with TensorFlow CPU runtime

Build serving with TensorFlow GPU runtime

Build serving with TensorFlow Lite CPU runtime

Build serving with TensorRT runtime

Build serving with TF-TRT runtime

Build serving with Tvm runtime

Build in Docker

Release

License