[Feature] [MLOps] support mlflow deploy with docker compose (#10217)

* [Feature] [MLOps] support mlflow deploy with docker compose

fix doc

Update docs/docs/en/guide/task/mlflow.md

fix doc

Co-authored-by: Jiajie Zhong <zhongjiajie955@gmail.com>

revert cancel modification

fix ENV name and docker compose command

* fix doc image link

* fix testModelsDeployDockerCompose

* add docker compose container health check and fix mlflow bug

* update docker compose healthcheck timeout

docs/docs/en/guide/task/mlflow.md

@@ -5,13 +5,13 @@
 [MLflow](https://mlflow.org) is an excellent open source platform to manage the ML lifecycle, including experimentation,
 reproducibility, deployment, and a central model registry.
 
-MLflow task plugin used to execute MLflow tasks，Currently contains Mlflow Projects and MLflow Models.（Model Registry will soon be rewarded for support）
+MLflow task plugin used to execute MLflow tasks，Currently contains MLflow Projects and MLflow Models. (Model Registry will soon be rewarded for support)
 
-- Mlflow Projects: Package data science code in a format to reproduce runs on any platform.
+- MLflow Projects: Package data science code in a format to reproduce runs on any platform.
 - MLflow Models: Deploy machine learning models in diverse serving environments.
 - Model Registry: Store, annotate, discover, and manage models in a central repository.
 
-The Mlflow plugin currently supports and will support the following:
+The MLflow plugin currently supports and will support the following:
 
 - [x] MLflow Projects
     - [x] BasicAlgorithm: contains LogisticRegression, svm, lightgbm, xgboost
@@ -20,10 +20,10 @@ The Mlflow plugin currently supports and will support the following:
 - [ ] MLflow Models
     - [x] MLFLOW: Use `MLflow models serve` to deploy a model service
     - [x] Docker: Run the container after packaging the docker image
-    - [ ] Docker Compose: Use docker compose to run the container, Will replace the docker run above
+    - [x] Docker Compose: Use docker compose to run the container, it will replace the docker run above
     - [ ] Seldon core: Use Selcon core to deploy model to k8s cluster
     - [ ] k8s: Deploy containers directly to K8S
-    - [ ] mlflow deployments: Built-in deployment modules, such as built-in deployment to SageMaker, etc
+    - [ ] MLflow deployments: Built-in deployment modules, such as built-in deployment to SageMaker, etc
 - [ ] Model Registry
     - [ ] Register Model: Allows artifacts (Including model and related parameters, indicators) to be registered directly into the model center
 
@@ -37,7 +37,7 @@ The Mlflow plugin currently supports and will support the following:
 
 ## Task Example
 
-First, introduce some general parameters of DolphinScheduler
+First, introduce some general parameters of DolphinScheduler:
 
 - **Node name**: The node name in a workflow definition is unique.
 - **Run flag**: Identifies whether this node schedules normally, if it does not need to execute, select
@@ -56,6 +56,11 @@ First, introduce some general parameters of DolphinScheduler
 - **Predecessor task**: Selecting a predecessor task for the current task, will set the selected predecessor task as
   upstream of the current task.
 
+Here are some specific parameters for the MLFlow component:
+
+- **MLflow Tracking Server URI**: MLflow Tracking Server URI, default http://localhost:5000.
+- **Experiment Name**: Create the experiment where the task is running, if the experiment does not exist. If the name is empty, it is set to ` Default `, the same as MLflow.
+
 ### MLflow Projects
 
 #### BasicAlgorithm
@@ -64,14 +69,12 @@ First, introduce some general parameters of DolphinScheduler
 
 **Task Parameter**
 
-- **mlflow server tracking uri** ：MLflow server uri, default http://localhost:5000.
+- **Register Model**: Register the model or not. If register is selected, the following parameters are expanded.
-- **experiment name** ：Create the experiment where the task is running, if the experiment does not exist. If the name is empty, it is set to ` Default `, the same as MLflow.
+    - **Model Name**: The registered model name is added to the original model version and registered as
-- **register model** ：Register the model or not. If register is selected, the following parameters are expanded.
-    - **model name** : The registered model name is added to the original model version and registered as
       Production.
-- **data path** : The absolute path of the file or folder. Ends with .csv for file or contain train.csv and
+- **Data Path**: The absolute path of the file or folder. Ends with .csv for file or contain train.csv and
-  test.csv for folder（In the suggested way, users should build their own test sets for model evaluation）。
+  test.csv for folder（In the suggested way, users should build their own test sets for model evaluation）.
-- **parameters** : Parameter when initializing the algorithm/AutoML model, which can be empty. For example
+- **Parameters**: Parameter when initializing the algorithm/AutoML model, which can be empty. For example
   parameters `"time_budget=30;estimator_list=['lgbm']"` for flaml 。The convention will be passed with '; ' shards
   each parameter, using the name before the equal sign as the parameter name, and using the name after the equal
   sign to get the corresponding parameter value through `python eval()`.
@@ -79,9 +82,9 @@ First, introduce some general parameters of DolphinScheduler
     - [SVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html?highlight=svc#sklearn.svm.SVC)
     - [lightgbm](https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.LGBMClassifier.html#lightgbm.LGBMClassifier)
     - [xgboost](https://xgboost.readthedocs.io/en/stable/python/python_api.html#xgboost.XGBClassifier)
-- **algorithm** ：The selected algorithm currently supports `LR`, `SVM`, `LightGBM` and `XGboost` based
+- **Algorithm**：The selected algorithm currently supports `LR`, `SVM`, `LightGBM` and `XGboost` based
   on [scikit-learn](https://scikit-learn.org/) form.
-- **Parameter search space** : Parameter search space when running the corresponding algorithm, which can be
+- **Parameter Search Space**: Parameter search space when running the corresponding algorithm, which can be
   empty. For example, the parameter `max_depth=[5, 10];n_estimators=[100, 200]` for lightgbm 。The convention
   will be passed with '; 'shards each parameter, using the name before the equal sign as the parameter name,
   and using the name after the equal sign to get the corresponding parameter value through `python eval()`.
@@ -92,63 +95,56 @@ First, introduce some general parameters of DolphinScheduler
 
 **Task Parameter**
 
-- **mlflow server tracking uri** ：MLflow server uri, default http://localhost:5000.
+- **Register Model**: Register the model or not. If register is selected, the following parameters are expanded.
-- **experiment name** ：Create the experiment where the task is running, if the experiment does not exist. If the name is empty, it is set to ` Default `, the same as MLflow.
-- **register model** ：Register the model or not. If register is selected, the following parameters are expanded.
     - **model name**: The registered model name is added to the original model version and registered as
       Production.
-- **data path** : The absolute path of the file or folder. Ends with .csv for file or contain train.csv and
+- **Data Path**: The absolute path of the file or folder. Ends with .csv for file or contain train.csv and
-  test.csv for folder（In the suggested way, users should build their own test sets for model evaluation）。
+  test.csv for folder(In the suggested way, users should build their own test sets for model evaluation).
-- **parameters** : Parameter when initializing the algorithm/AutoML model, which can be empty. For example
+- **Parameters**: Parameter when initializing the algorithm/AutoML model, which can be empty. For example
-  parameters `n_estimators=200;learning_rate=0.2` for flaml 。The convention will be passed with '; 'shards
+  parameters `n_estimators=200;learning_rate=0.2` for flaml. The convention will be passed with '; 'shards
   each parameter, using the name before the equal sign as the parameter name, and using the name after the equal
   sign to get the corresponding parameter value through `python eval()`. The detailed parameter list is as follows:
   - [flaml](https://microsoft.github.io/FLAML/docs/reference/automl#automl-objects)
   - [autosklearn](https://automl.github.io/auto-sklearn/master/api.html)
 - **AutoML tool**: The AutoML tool used, currently
   supports [autosklearn](https://github.com/automl/auto-sklearn)
-  and [flaml](https://github.com/microsoft/FLAML)
+  and [flaml](https://github.com/microsoft/FLAML).
-
 
 #### Custom projects
 
-![mlflow-custom-project-template.png](../../../../img/tasks/demo/mlflow-custom-project-template.png)
+![mlflow-custom-project.png](../../../../img/tasks/demo/mlflow-custom-project.png)
 
 **Task Parameter**
 
-- **mlflow server tracking uri** ：MLflow server uri, default http://localhost:5000.
+- **parameters**: `--param-list` in `mlflow run`. For example `-P learning_rate=0.2 -P colsample_bytree=0.8 -P subsample=0.9`.
-- **experiment name** ：Create the experiment where the task is running, if the experiment does not exist. If the name is empty, it is set to ` Default `, the same as MLflow.
+- **Repository**: Repository url of MLflow Project，Support git address and directory on worker. If it's in a subdirectory，We add `#` to support this (same as `mlflow run`) , for example `https://github.com/mlflow/mlflow#examples/xgboost/xgboost_native`.
-- **parameters** : `--param-list` in `mlflow run`. For example `-P learning_rate=0.2 -P colsample_bytree=0.8 -P subsample=0.9`
+- **Project Version**: Version of the project，default master.
-- **Repository** : Repository url of MLflow Project，Support git address and directory on worker. If it's in a subdirectory，We add `#` to support this (same as `mlflow run`) , for example `https://github.com/mlflow/mlflow#examples/xgboost/xgboost_native`
-- **Project Version** : Version of the project，default master
-
-You can now use this feature to run all mlFlow projects on Github (For example [MLflow examples](https://github.com/mlflow/mlflow/tree/master/examples) )了。You can also create your own machine learning library to reuse your work, and then use DolphinScheduler to use your library with one click.
 
-The actual interface is as follows
+You can now use this feature to run all MLFlow projects on Github (For example [MLflow examples](https://github.com/mlflow/mlflow/tree/master/examples) ). You can also create your own machine learning library to reuse your work, and then use DolphinScheduler to use your library with one click.
 
-![mlflow-custom-project.png](../../../../img/tasks/demo/mlflow-custom-project.png)
 
 ### MLflow Models
 
-#### MLFLOW
+General Parameters:
 
-![mlflow-models-mlflow](../../../../img/tasks/demo/mlflow-models-mlflow.png)
+- **Model-URI**: Model-URI of MLflow , support `models:/<model_name>/suffix` format and `runs:/` format. See https://mlflow.org/docs/latest/tracking.html#artifact-stores.
+- **Port**: The port to listen on.
 
-**Task Parameter**
 
-- **mlflow server tracking uri** ：MLflow server uri, default http://localhost:5000.
+#### MLFLOW
-- **model-uri** ：Model-uri of mlflow , support `models:/<model_name>/suffix` format and `runs:/` format. See https://mlflow.org/docs/latest/tracking.html#artifact-stores
+
-- **Port** ：The port to listen on
+![mlflow-models-mlflow](../../../../img/tasks/demo/mlflow-models-mlflow.png)
 
 #### Docker
 
 ![mlflow-models-docker](../../../../img/tasks/demo/mlflow-models-docker.png)
 
-**Task Parameter**
+#### DOCKER COMPOSE
+
+![mlflow-models-docker-compose](../../../../img/tasks/demo/mlflow-models-docker-compose.png)
 
-- **mlflow server tracking uri** ：MLflow server uri, default http://localhost:5000.
+- **Max Cpu Limit**: For example `1.0` or `0.5`, the same as docker compose.
-- **model-uri** ：Model-uri of mlflow , support `models:/<model_name>/suffix` format and `runs:/` format. See https://mlflow.org/docs/latest/tracking.html#artifact-stores
+- **Max Memory Limit**: For example `1G` or `500M`, the same as docker compose.
-- **Port** ：The port to listen on
 
 ## Environment to prepare
 
@@ -156,7 +152,7 @@ The actual interface is as follows
 
 You need to enter the admin account to configure a conda environment variable（Please
 install [anaconda](https://docs.continuum.io/anaconda/install/)
-or [miniconda](https://docs.conda.io/en/latest/miniconda.html#installing ) in advance )
+or [miniconda](https://docs.conda.io/en/latest/miniconda.html#installing ) in advance).
 
 ![mlflow-conda-env](../../../../img/tasks/demo/mlflow-conda-env.png)
 
@@ -167,9 +163,9 @@ Conda environment.
 
 ### Start the mlflow service
 
-Make sure you have installed MLflow, using 'PIP Install MLFlow'.
+Make sure you have installed MLflow, using 'pip install mlflow'.
 
-Create a folder where you want to save your experiments and models and start mlFlow service.
+Create a folder where you want to save your experiments and models and start MLflow service.
 
 ```sh
 mkdir mlflow
@@ -177,8 +173,8 @@ cd mlflow
 mlflow server -h 0.0.0.0 -p 5000 --serve-artifacts --backend-store-uri sqlite:///mlflow.db
 ```
 
-After running, an MLflow service is started
+After running, an MLflow service is started.
 
-After this, you can visit the MLFlow service (`http://localhost:5000`) page to view the experiments and models.
+After this, you can visit the MLflow service (`http://localhost:5000`) page to view the experiments and models.
 
 ![mlflow-server](../../../../img/tasks/demo/mlflow-server.png)

docs/docs/zh/guide/task/mlflow.md

@@ -4,25 +4,25 @@
 
 [MLflow](https://mlflow.org) 是一个MLops领域一个优秀的开源项目， 用于管理机器学习的生命周期，包括实验、可再现性、部署和中心模型注册。
 
-MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow Models。（Model Registry将在不就的将来支持）
+MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow Models。（Model Registry将在不就的将来支持）。
 
-- Mlflow Projects: 将代码打包，并可以运行到任务的平台上。
+- MLflow Projects: 将代码打包，并可以运行到任务的平台上。
 - MLflow Models: 在不同的服务环境中部署机器学习模型。
-- Model Registry: 在一个中央存储库中存储、注释、发现和管理模型 (你也可以在你的mlflow project 里面自行注册模型)。
+- Model Registry: 在一个中央存储库中存储、注释、发现和管理模型 (你也可以在你的MLflow project 里面自行注册模型)。
 
 目前 Mlflow 组件支持的和即将支持的内容如下中：
 
 - [x] MLflow Projects
-  - [x] BasicAlgorithm: 基础算法，包含LogisticRegression, svm, lightgbm, xgboost。
+  - [x] BasicAlgorithm: 基础算法，包含LogisticRegression, svm, lightgbm, xgboost
-  - [x] AutoML: AutoML工具，包含autosklean, flaml。
+  - [x] AutoML: AutoML工具，包含autosklean, flaml
   - [x] Custom projects: 支持运行自己的MLflow Projects项目
 - [ ] MLflow Models
-  - [x] MLFLOW: 直接使用 `MLflow models serve` 部署模型
+  - [x] MLFLOW: 直接使用 `mlflow models serve` 部署模型。
-  - [x] Docker: 打包 DOCKER 镜像后部署模型
+  - [x] Docker: 打包 DOCKER 镜像后部署模型。
-  - [ ] Docker Compose: 使用Docker Compose 部署模型，将会取代上面的Docker部署
+  - [x] Docker Compose: 使用Docker Compose 部署模型，将会取代上面的Docker部署。
-  - [ ] Seldon core: 构建完镜像后，使用Seldon Core 部署到k8s集群上, 可以使用Seldon Core的生成模型管理能力
+  - [ ] Seldon core: 构建完镜像后，使用Seldon Core 部署到k8s集群上, 可以使用Seldon Core的生成模型管理能力。
-  - [ ] k8s: 构建完镜像后， 部署到k8s集群上
+  - [ ] k8s: 构建完镜像后， 部署到k8s集群上。
-  - [ ] mlflow deployments: 内置的允许MLflow 部署模块, 如内置的部署到Sagemaker等
+  - [ ] MLflow deployments: 内置的允许MLflow 部署模块, 如内置的部署到Sagemaker等。
 - [ ] Model Registry
   - [ ] Register Model: 注册相关工件(模型以及相关的参数，指标)到模型中心
 
@@ -48,6 +48,12 @@ MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow
 - **超时告警** ：勾选超时告警、超时失败，当任务超过"超时时长"后，会发送告警邮件并且任务执行失败。
 - **前置任务** ：选择当前任务的前置任务，会将被选择的前置任务设置为当前任务的上游。
 
+
+以下是一些MLflow 组件的常用参数
+
+- **MLflow Tracking Server URI** ：MLflow Tracking Server 的连接, 默认 http://localhost:5000。
+- **实验名称** ：任务运行时所在的实验，若实验不存在，则创建。若实验名称为空，则设置为`Default`, 与 MLflow 一样。
+
 ### MLflow Projects
 
 #### BasicAlgorithm
@@ -56,8 +62,6 @@ MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow
 
 **任务参数**
 
-- **mlflow server tracking uri** ：MLflow server 的连接, 默认 http://localhost:5000。
-- **实验名称** ：任务运行时所在的实验，若实验不存在，则创建。若实验名称为空，则设置为`Default`, 与 MLflow 一样。
 - **注册模型** ：是否注册模型，若选择注册，则会展开以下参数。
     - **注册的模型名称** : 注册的模型名称，会在原来的基础上加上一个模型版本，并注册为Production。
 - **数据路径** : 文件/文件夹的绝对路径, 若文件需以.csv结尾（自动切分训练集与测试集）, 文件夹需包含train.csv和test.csv（建议方式，用户应自行构建测试集用于模型评估）。
@@ -66,7 +70,7 @@ MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow
   - [SVM](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html?highlight=svc#sklearn.svm.SVC)
   - [lightgbm](https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.LGBMClassifier.html#lightgbm.LGBMClassifier)
   - [xgboost](https://xgboost.readthedocs.io/en/stable/python/python_api.html#xgboost.XGBClassifier)
-- **算法** ：选择的算法，目前基于 [scikit-learn](https://scikit-learn.org/) 形式支持 `lr`, `svm`, `lightgbm`, `xgboost`.
+- **算法** ：选择的算法，目前基于 [scikit-learn](https://scikit-learn.org/) 形式支持 `lr`, `svm`, `lightgbm`, `xgboost`。
 - **参数搜索空间** : 运行对应算法的参数搜索空间, 可为空。如针对lightgbm 的 `max_depth=[5, 10];n_estimators=[100, 200]` 则会进行对应搜索。约定传入后会以;切分各个参数，等号前的名字作为参数名，等号后的名字将以python eval执行得到对应的参数值
 
 #### AutoML
@@ -75,8 +79,6 @@ MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow
 
 **任务参数**
 
-- **mlflow server tracking uri** ：MLflow server 的连接, 默认 http://localhost:5000。
-- **实验名称** ：任务运行时所在的实验，若实验不存在，则创建。若实验名称为空，则设置为`Default`, 与 MLflow 一样。
 - **注册模型** ：是否注册模型，若选择注册，则会展开以下参数。
     - **注册的模型名称** : 注册的模型名称，会在原来的基础上加上一个模型版本，并注册为Production。
 - **数据路径** : 文件/文件夹的绝对路径, 若文件需以.csv结尾（自动切分训练集与测试集）, 文件夹需包含train.csv和test.csv（建议方式，用户应自行构建测试集用于模型评估）。
@@ -84,65 +86,61 @@ MLflow 组件用于执行 MLflow 任务，目前包含Mlflow Projects, 和MLflow
   - [flaml](https://microsoft.github.io/FLAML/docs/reference/automl#automl-objects)
   - [autosklearn](https://automl.github.io/auto-sklearn/master/api.html)
 - **AutoML工具** : 使用的AutoML工具，目前支持 [autosklearn](https://github.com/automl/auto-sklearn)
-  , [flaml](https://github.com/microsoft/FLAML)
+  , [flaml](https://github.com/microsoft/FLAML)。
 
 #### Custom projects
 
-![mlflow-custom-project-template.png](../../../../img/tasks/demo/mlflow-custom-project-template.png)
+![mlflow-custom-project.png](../../../../img/tasks/demo/mlflow-custom-project.png)
 
 **任务参数**
 
-- **mlflow server tracking uri** ：MLflow server 的连接, 默认 http://localhost:5000。
-- **实验名称** ：任务运行时所在的实验，若实验不存在，则创建。若实验名称为空，则设置为`Default`, 与 MLflow 一样。
 - **参数** : `mlflow run`中的 --param-list 如 `-P learning_rate=0.2 -P colsample_bytree=0.8 -P subsample=0.9`
-- **运行仓库** : MLflow Project的仓库地址，可以为github地址，或者worker上的目录, 如Mlflow project位于子目录，可以添加 `#` 隔开, 如 `https://github.com/mlflow/mlflow#examples/xgboost/xgboost_native`
+- **运行仓库** : MLflow Project的仓库地址，可以为github地址，或者worker上的目录, 如MLflow project位于子目录，可以添加 `#` 隔开, 如 `https://github.com/mlflow/mlflow#examples/xgboost/xgboost_native`
 - **项目版本** : 对应项目中git版本管理中的版本，默认 master
 
 现在你可以使用这个功能来运行github上所有的MLflow Projects (如 [MLflow examples](https://github.com/mlflow/mlflow/tree/master/examples) )了。你也可以创建自己的机器学习库，用来复用你的研究成果，以后你就可以使用DolphinScheduler来一键操作使用你的算法库。
 
-实际运行界面如下
 
-![mlflow-custom-project.png](../../../../img/tasks/demo/mlflow-custom-project.png)
+### MLflow Models
 
+常用参数:
 
-### MLflow Models
+- **部署模型的URI** ：MLflow 服务里面模型对应的URI, 支持 `models:/<model_name>/suffix` 格式 和 `runs:/` 格式。
+- **监听端口** ：部署服务时的端口。
 
 #### MLFLOW
 
 ![mlflow-models-mlflow](../../../../img/tasks/demo/mlflow-models-mlflow.png)
 
-**任务参数**
-
-- **mlflow server tracking uri** ：MLflow server 的连接, 默认 http://localhost:5000。
-- **部署模型的uri** ：mlflow 服务里面模型对应的uri, 支持 `models:/<model_name>/suffix` 格式 和 `runs:/` 格式。
-- **部署端口** ：部署服务时的端口。
-
 #### Docker
 
 ![mlflow-models-docker](../../../../img/tasks/demo/mlflow-models-docker.png)
 
-- **mlflow server tracking uri** ：MLflow server 的连接, 默认 http://localhost:5000。
+#### DOCKER COMPOSE
-- **部署模型的uri** ：mlflow 服务里面模型对应的uri, 支持 `models:/<model_name>/suffix` 格式 和 `runs:/` 格式。
+
-- **部署端口** ：部署服务时的端口。
+![mlflow-models-docker-compose](../../../../img/tasks/demo/mlflow-models-docker-compose.png)
+
+- **最大CPU限制** ：如 `1.0` 或者 `0.5`, 与 docker compose 一致。
+- **最大内存限制** ：如 `1G` 或者 `500M`, 与 docker compose 一致。
 
 ## 环境准备
 
 ### conda 环境配置
 
 你需要进入admin账户配置一个conda环境变量（请提前[安装anaconda](https://docs.continuum.io/anaconda/install/)
-或者[安装miniconda](https://docs.conda.io/en/latest/miniconda.html#installing) )
+或者[安装miniconda](https://docs.conda.io/en/latest/miniconda.html#installing) )。
 
 ![mlflow-conda-env](../../../../img/tasks/demo/mlflow-conda-env.png)
 
-后续注意配置任务时，环境选择上面创建的conda环境，否则程序会找不到conda环境
+后续注意配置任务时，环境选择上面创建的conda环境，否则程序会找不到conda环境。
 
 ![mlflow-set-conda-env](../../../../img/tasks/demo/mlflow-set-conda-env.png)
 
-### mlflow service 启动
+### MLflow service 启动
 
-确保你已经安装mlflow，可以使用`pip install mlflow`进行安装
+确保你已经安装MLflow，可以使用`pip install mlflow`进行安装。
 
-在你想保存实验和模型的地方建立一个文件夹，然后启动 mlflow service
+在你想保存实验和模型的地方建立一个文件夹，然后启动 mlflow service。
 
 ```sh
 mkdir mlflow
@@ -150,9 +148,9 @@ cd mlflow
 mlflow server -h 0.0.0.0 -p 5000 --serve-artifacts --backend-store-uri sqlite:///mlflow.db
 ```
 
-运行后会启动一个mlflow服务
+运行后会启动一个MLflow服务。
 
-可以通过访问 mlflow service (`http://localhost:5000`) 页面查看实验与模型
+可以通过访问 MLflow service (`http://localhost:5000`) 页面查看实验与模型。
 
 ![mlflow-server](../../../../img/tasks/demo/mlflow-server.png)
 

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/main/java/org/apache/dolphinscheduler/plugin/task/mlflow/MlflowConstants.java

@@ -36,19 +36,21 @@ public class MlflowConstants {
 
     public static final String PRESET_BASIC_ALGORITHM_PROJECT = PRESET_REPOSITORY + "#Project-BasicAlgorithm";
 
-    public static final String RUN_PROJECT_BASIC_ALGORITHM_SCRIPT = "run_mlflow_basic_algorithm_project.sh";
-
-    public static final String RUN_PROJECT_AUTOML_SCRIPT = "run_mlflow_automl_project.sh";
-
     public static final String MLFLOW_TASK_TYPE_PROJECTS = "MLflow Projects";
 
     public static final String MLFLOW_TASK_TYPE_MODELS = "MLflow Models";
 
-
     public static final String MLFLOW_MODELS_DEPLOY_TYPE_MLFLOW = "MLFLOW";
 
     public static final String MLFLOW_MODELS_DEPLOY_TYPE_DOCKER = "DOCKER";
 
+    public static final String MLFLOW_MODELS_DEPLOY_TYPE_DOCKER_COMPOSE = "DOCKER COMPOSE";
+
+    /**
+     * template file
+     */
+    public static final String TEMPLATE_DOCKER_COMPOSE = "docker-compose.yml";
+
 
     /**
      * mlflow command
@@ -86,9 +88,22 @@ public class MlflowConstants {
 
     public static final String MLFLOW_BUILD_DOCKER = "mlflow models build-docker -m %s -n %s --enable-mlserver";
 
-
     public static final String DOCKER_RREMOVE_CONTAINER = "docker rm -f %s";
 
     public static final String DOCKER_RUN = "docker run --name=%s -p=%s:8080 %s";
 
+    public static final String DOCKER_COMPOSE_RUN = "docker-compose up -d";
+
+    public static final String SET_DOCKER_COMPOSE_ENV = "export DS_TASK_MLFLOW_IMAGE_NAME=%s\n" +
+            "export DS_TASK_MLFLOW_CONTAINER_NAME=%s\n" +
+            "export DS_TASK_MLFLOW_DEPLOY_PORT=%s\n" +
+            "export DS_TASK_MLFLOW_CPU_LIMIT=%s\n" +
+            "export DS_TASK_MLFLOW_MEMORY_LIMIT=%s";
+
+    public static final String DOCKER_HEALTH_CHECK_COMMAND = "for i in $(seq 1 300); " +
+            "do " +
+            "[ $(docker inspect --format \"{{json .State.Health.Status }}\" %s) = '\"healthy\"' ] " +
+            "&& exit 0  && break;sleep 1; " +
+            "done; docker-compose down; exit 1";
+
 }

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/main/java/org/apache/dolphinscheduler/plugin/task/mlflow/MlflowParameters.java

@@ -76,6 +76,10 @@ public class MlflowParameters extends AbstractParameters {
 
     private String deployPort;
 
+    private String cpuLimit;
+
+    private String memoryLimit;
+
     public void setAlgorithm(String algorithm) {
         this.algorithm = algorithm;
     }
@@ -196,6 +200,22 @@ public class MlflowParameters extends AbstractParameters {
         return deployPort;
     }
 
+    public void setCpuLimit(String cpuLimit) {
+        this.cpuLimit = cpuLimit;
+    }
+
+    public String getCpuLimit() {
+        return cpuLimit;
+    }
+
+    public void setMemoryLimit(String memoryLimit) {
+        this.memoryLimit = memoryLimit;
+    }
+
+    public String getMemoryLimit() {
+        return memoryLimit;
+    }
+
     @Override
     public boolean checkParameters() {
         Boolean checkResult = true;
@@ -242,19 +262,6 @@ public class MlflowParameters extends AbstractParameters {
         paramsMap.put("repo_version", MlflowConstants.PRESET_REPOSITORY_VERSION);
     }
 
-    public String getScriptPath() {
-        String projectScript;
-        if (mlflowJobType.equals(MlflowConstants.JOB_TYPE_BASIC_ALGORITHM)) {
-            projectScript = MlflowConstants.RUN_PROJECT_BASIC_ALGORITHM_SCRIPT;
-        } else if (mlflowJobType.equals(MlflowConstants.JOB_TYPE_AUTOML)) {
-            projectScript = MlflowConstants.RUN_PROJECT_AUTOML_SCRIPT;
-        } else {
-            throw new IllegalArgumentException();
-        }
-        String scriptPath = MlflowTask.class.getClassLoader().getResource(projectScript).getPath();
-        return scriptPath;
-    }
-
     public String getModelKeyName(String tag) throws IllegalArgumentException {
         String imageName;
         if (deployModelKey.startsWith("runs:")) {
@@ -268,4 +275,15 @@ public class MlflowParameters extends AbstractParameters {
         return imageName;
     }
 
+    public String getDockerComposeEnvCommand() {
+        String imageName = "mlflow/" + getModelKeyName(":");
+        String env = String.format(MlflowConstants.SET_DOCKER_COMPOSE_ENV, imageName, getContainerName(), deployPort, cpuLimit, memoryLimit);
+        return env;
+    }
+
+    public String getContainerName(){
+        String containerName = "ds-mlflow-" + getModelKeyName("-");
+        return containerName;
+    }
+
 };

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/main/java/org/apache/dolphinscheduler/plugin/task/mlflow/MlflowTask.java

@@ -146,8 +146,7 @@ public class MlflowTask extends AbstractTaskExecutor {
 
             runCommand = MlflowConstants.MLFLOW_RUN_CUSTOM_PROJECT;
             runCommand = String.format(runCommand, mlflowParameters.getParams(), mlflowParameters.getExperimentName(), mlflowParameters.getMlflowProjectVersion());
-        }
+        } else {
-        else {
             runCommand = String.format("Cant not Support %s", mlflowParameters.getMlflowJobType());
 
         }
@@ -173,11 +172,19 @@ public class MlflowTask extends AbstractTaskExecutor {
 
         } else if (mlflowParameters.getDeployType().equals(MlflowConstants.MLFLOW_MODELS_DEPLOY_TYPE_DOCKER)) {
             String imageName = "mlflow/" + mlflowParameters.getModelKeyName(":");
-            String containerName = "mlflow-" + mlflowParameters.getModelKeyName("-");
+            String containerName = mlflowParameters.getContainerName();
 
             args.add(String.format(MlflowConstants.MLFLOW_BUILD_DOCKER, deployModelKey, imageName));
             args.add(String.format(MlflowConstants.DOCKER_RREMOVE_CONTAINER, containerName));
             args.add(String.format(MlflowConstants.DOCKER_RUN, containerName, mlflowParameters.getDeployPort(), imageName));
+        } else if (mlflowParameters.getDeployType().equals(MlflowConstants.MLFLOW_MODELS_DEPLOY_TYPE_DOCKER_COMPOSE)) {
+            String templatePath = getTemplatePath(MlflowConstants.TEMPLATE_DOCKER_COMPOSE);
+            args.add(String.format("cp %s %s", templatePath, taskExecutionContext.getExecutePath()));
+            String imageName = "mlflow/" + mlflowParameters.getModelKeyName(":");
+            args.add(String.format(MlflowConstants.MLFLOW_BUILD_DOCKER, deployModelKey, imageName));
+            args.add(mlflowParameters.getDockerComposeEnvCommand());
+            args.add(MlflowConstants.DOCKER_COMPOSE_RUN);
+            args.add(String.format(MlflowConstants.DOCKER_HEALTH_CHECK_COMMAND, mlflowParameters.getContainerName()));
         }
 
         String command = ParameterUtils.convertParameterPlaceholders(String.join("\n", args), ParamUtils.convert(paramsMap));
@@ -197,9 +204,15 @@ public class MlflowTask extends AbstractTaskExecutor {
 
     }
 
+
     @Override
     public AbstractParameters getParameters() {
         return mlflowParameters;
     }
 
+    public String getTemplatePath(String template) {
+        String templatePath = MlflowTask.class.getClassLoader().getResource(template).getPath();
+        return templatePath;
+    }
+
 }

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/main/resources/docker-compose.yml

@@ -0,0 +1,39 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+version: "3"
+
+services:
+  mlflow-model:
+    image: "${DS_TASK_MLFLOW_IMAGE_NAME}"
+    container_name: "${DS_TASK_MLFLOW_CONTAINER_NAME}"
+    ports:
+      - "${DS_TASK_MLFLOW_DEPLOY_PORT}:8080"
+    deploy:
+      resources:
+        limits:
+          cpus: "${DS_TASK_MLFLOW_CPU_LIMIT}"
+          memory: "${DS_TASK_MLFLOW_MEMORY_LIMIT}"
+
+    environment:
+      PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION: python
+
+
+    healthcheck:
+      test: ["CMD", "curl", "http://127.0.0.1:8080/ping"]
+      interval: 5s
+      timeout: 5s
+      retries: 5

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/main/resources/run_mlflow_automl_project.sh

@@ -1,25 +0,0 @@
-#!/bin/bash
-#
-# Licensed to the Apache Software Foundation (ASF) under one or more
-# contributor license agreements.  See the NOTICE file distributed with
-# this work for additional information regarding copyright ownership.
-# The ASF licenses this file to You under the Apache License, Version 2.0
-# (the "License"); you may not use this file except in compliance with
-# the License.  You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-
-data_path=${data_path}
-export MLFLOW_TRACKING_URI=${MLFLOW_TRACKING_URI}
-echo $data_path
-repo=${repo}
-mlflow run $repo -P tool=${automl_tool} -P data_path=$data_path -P params="${params}" -P model_name="${model_name}" --experiment-name="${experiment_name}" --version="${repo_version}"
-
-echo "training finish"

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/main/resources/run_mlflow_basic_algorithm_project.sh

@@ -1,25 +0,0 @@
-#!/bin/bash
-#
-# Licensed to the Apache Software Foundation (ASF) under one or more
-# contributor license agreements.  See the NOTICE file distributed with
-# this work for additional information regarding copyright ownership.
-# The ASF licenses this file to You under the Apache License, Version 2.0
-# (the "License"); you may not use this file except in compliance with
-# the License.  You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-
-data_path=${data_path}
-export MLFLOW_TRACKING_URI=${MLFLOW_TRACKING_URI}
-echo $data_path
-repo=${repo}
-mlflow run $repo -P algorithm=${algorithm} -P data_path=$data_path -P params="${params}" -P search_params="${search_params}" -P model_name="${model_name}" --experiment-name="${experiment_name}" --version="${repo_version}"
-
-echo "training finish"

dolphinscheduler-task-plugin/dolphinscheduler-task-mlflow/src/test/java/org/apache/dolphinler/plugin/task/mlflow/MlflowTaskTest.java

@@ -135,18 +135,37 @@ public class MlflowTaskTest {
     }
 
     @Test
-    public void testModelsDeployDocker() throws Exception {
+    public void testModelsDeployDocker() {
         MlflowTask mlflowTask = initTask(createModelDeplyDockerParameters());
         Assert.assertEquals(mlflowTask.buildCommand(),
                 "export MLFLOW_TRACKING_URI=http://127.0.0.1:5000\n" +
                         "mlflow models build-docker -m runs:/a272ec279fc34a8995121ae04281585f/model " +
                         "-n mlflow/a272ec279fc34a8995121ae04281585f:model " +
                         "--enable-mlserver\n" +
-                        "docker rm -f mlflow-a272ec279fc34a8995121ae04281585f-model\n" +
+                        "docker rm -f ds-mlflow-a272ec279fc34a8995121ae04281585f-model\n" +
-                        "docker run --name=mlflow-a272ec279fc34a8995121ae04281585f-model " +
+                        "docker run --name=ds-mlflow-a272ec279fc34a8995121ae04281585f-model " +
                         "-p=7000:8080 mlflow/a272ec279fc34a8995121ae04281585f:model");
     }
 
+    @Test
+    public void testModelsDeployDockerCompose() throws Exception{
+        MlflowTask mlflowTask = initTask(createModelDeplyDockerComposeParameters());
+        Assert.assertEquals(mlflowTask.buildCommand(),
+                "export MLFLOW_TRACKING_URI=http://127.0.0.1:5000\n" +
+                        "cp " + mlflowTask.getTemplatePath(MlflowConstants.TEMPLATE_DOCKER_COMPOSE) +
+                        " /tmp/dolphinscheduler_test\n" +
+                        "mlflow models build-docker -m models:/22222/1 -n mlflow/22222:1 --enable-mlserver\n" +
+                        "export DS_TASK_MLFLOW_IMAGE_NAME=mlflow/22222:1\n" +
+                        "export DS_TASK_MLFLOW_CONTAINER_NAME=ds-mlflow-22222-1\n" +
+                        "export DS_TASK_MLFLOW_DEPLOY_PORT=7000\n" +
+                        "export DS_TASK_MLFLOW_CPU_LIMIT=0.5\n" +
+                        "export DS_TASK_MLFLOW_MEMORY_LIMIT=200m\n" +
+                        "docker-compose up -d\n" +
+                        "for i in $(seq 1 300); do " +
+                        "[ $(docker inspect --format \"{{json .State.Health.Status }}\" ds-mlflow-22222-1) = '\"healthy\"' ] && exit 0  && break;sleep 1; " +
+                        "done; docker-compose down; exit 1");
+    }
+
     private MlflowTask initTask(MlflowParameters mlflowParameters) {
         TaskExecutionContext taskExecutionContext = createContext(mlflowParameters);
         MlflowTask mlflowTask = new MlflowTask(taskExecutionContext);
@@ -213,4 +232,16 @@ public class MlflowTaskTest {
         mlflowParameters.setDeployPort("7000");
         return mlflowParameters;
     }
+
+    private MlflowParameters createModelDeplyDockerComposeParameters() {
+        MlflowParameters mlflowParameters = new MlflowParameters();
+        mlflowParameters.setMlflowTaskType(MlflowConstants.MLFLOW_TASK_TYPE_MODELS);
+        mlflowParameters.setDeployType(MlflowConstants.MLFLOW_MODELS_DEPLOY_TYPE_DOCKER_COMPOSE);
+        mlflowParameters.setMlflowTrackingUris("http://127.0.0.1:5000");
+        mlflowParameters.setDeployModelKey("models:/22222/1");
+        mlflowParameters.setDeployPort("7000");
+        mlflowParameters.setCpuLimit("0.5");
+        mlflowParameters.setMemoryLimit("200m");
+        return mlflowParameters;
+    }
 }

dolphinscheduler-ui/src/locales/en_US/project.ts

@@ -608,9 +608,6 @@ export default {
     zeppelin_paragraph_id: 'zeppelinParagraphId',
     zeppelin_paragraph_id_tips:
       'Please enter the paragraph id of your zeppelin paragraph',
-    zeppelin_parameters: 'parameters',
-    zeppelin_parameters_tips:
-      'Please enter the parameters for zeppelin dynamic form',
     jupyter_conda_env_name: 'condaEnvName',
     jupyter_conda_env_name_tips:
       'Please enter the conda environment name of papermill',
@@ -634,36 +631,38 @@ export default {
     jupyter_others: 'others',
     jupyter_others_tips:
       'Please enter the other options you need for papermill',
-    mlflow_algorithm: 'algorithm',
+    mlflow_algorithm: 'Algorithm',
     mlflow_algorithm_tips: 'svm',
-    mlflow_params: 'parameters',
+    mlflow_params: 'Parameters',
     mlflow_params_tips: ' ',
-    mlflow_searchParams: 'Parameter search space',
+    mlflow_searchParams: 'Parameter Search Space',
     mlflow_searchParams_tips: ' ',
-    mlflow_isSearchParams: 'Search parameters',
+    mlflow_isSearchParams: 'Search Parameters',
-    mlflow_dataPath: 'data path',
+    mlflow_dataPath: 'Data Path',
     mlflow_dataPath_tips:
       ' The absolute path of the file or folder. Ends with .csv for file or contain train.csv and test.csv for folder',
     mlflow_dataPath_error_tips: ' data data can not be empty ',
-    mlflow_experimentName: 'experiment name',
+    mlflow_experimentName: 'Experiment Name',
     mlflow_experimentName_tips: 'experiment_001',
-    mlflow_registerModel: 'register model',
+    mlflow_registerModel: 'Register Model',
-    mlflow_modelName: 'model name',
+    mlflow_modelName: 'Model Name',
     mlflow_modelName_tips: 'model_001',
-    mlflow_mlflowTrackingUri: 'mlflow server tracking uri',
+    mlflow_mlflowTrackingUri: 'MLflow Tracking Server URI',
     mlflow_mlflowTrackingUri_tips: 'http://127.0.0.1:5000',
     mlflow_mlflowTrackingUri_error_tips:
-      ' mlflow server tracking uri cant not be empty',
+      'MLflow Tracking Server URI can not be empty',
-    mlflow_jobType: 'job type',
+    mlflow_jobType: 'Job Type',
-    mlflow_automlTool: 'AutoML tool',
+    mlflow_automlTool: 'AutoML Tool',
     mlflow_taskType: 'MLflow Task Type',
     mlflow_deployType: 'Deploy Mode',
-    mlflow_deployModelKey: 'model-uri',
+    mlflow_deployModelKey: 'Model-URI',
     mlflow_deployPort: 'Port',
     mlflowProjectRepository: 'Repository',
     mlflowProjectRepository_tips: 'github respository or path on worker',
     mlflowProjectVersion: 'Project Version',
     mlflowProjectVersion_tips: 'git version',
+    mlflow_cpuLimit: 'Max Cpu Limit',
+    mlflow_memoryLimit: 'Max Memory Limit',
     openmldb_zk_address: 'zookeeper address',
     openmldb_zk_address_tips: 'Please enter the zookeeper address',
     openmldb_zk_path: 'zookeeper path',

dolphinscheduler-ui/src/locales/zh_CN/project.ts

@@ -637,19 +637,21 @@ export default {
     mlflow_registerModel: '注册模型',
     mlflow_modelName: '注册的模型名称',
     mlflow_modelName_tips: 'model_001',
-    mlflow_mlflowTrackingUri: 'mlflow server tracking uri',
+    mlflow_mlflowTrackingUri: 'MLflow Tracking Server URI',
     mlflow_mlflowTrackingUri_tips: 'http://127.0.0.1:5000',
-    mlflow_mlflowTrackingUri_error_tips: ' mlflow server tracking uri 不能为空',
+    mlflow_mlflowTrackingUri_error_tips: ' MLflow Tracking Server URI 不能为空',
     mlflow_jobType: '任务类型',
     mlflow_automlTool: 'AutoML工具',
     mlflow_taskType: 'MLflow 任务类型',
     mlflow_deployType: '部署类型',
-    mlflow_deployModelKey: '部署的模型uri',
+    mlflow_deployModelKey: '部署的模型URI',
     mlflow_deployPort: '监听端口',
     mlflowProjectRepository: '运行仓库',
     mlflowProjectRepository_tips: '可以为github仓库或worker上的路径',
     mlflowProjectVersion: '项目版本',
     mlflowProjectVersion_tips: '项目git版本',
+    mlflow_cpuLimit: '最大cpu限制',
+    mlflow_memoryLimit: '最大内存限制',
     openmldb_zk_address: 'zookeeper地址',
     openmldb_zk_address_tips: '请输入zookeeper地址',
     openmldb_zk_path: 'zookeeper路径',

dolphinscheduler-ui/src/views/projects/task/components/node/fields/use-mlflow-models.ts

@@ -23,6 +23,8 @@ export function useMlflowModels(model: { [field: string]: any }): IJsonItem[] {
   const deployTypeSpan = ref(0)
   const deployModelKeySpan = ref(0)
   const deployPortSpan = ref(0)
+  const cpuLimitSpan = ref(0)
+  const memoryLimitSpan = ref(0)
 
   const setFlag = () => {
     model.isModels = model.mlflowTaskType === 'MLflow Models' ? true : false
@@ -35,13 +37,21 @@ export function useMlflowModels(model: { [field: string]: any }): IJsonItem[] {
   }
 
   watch(
-    () => [model.mlflowTaskType, model.registerModel],
+    () => [model.mlflowTaskType],
     () => {
       setFlag()
       resetSpan()
     }
   )
 
+  watch(
+    () => [model.deployType],
+    () => {
+     cpuLimitSpan.value = model.deployType === "DOCKER COMPOSE" ? 12 : 0 
+     memoryLimitSpan.value = model.deployType === "DOCKER COMPOSE" ? 12 : 0 
+    }
+  )
+
   setFlag()
   resetSpan()
 
@@ -64,6 +74,18 @@ export function useMlflowModels(model: { [field: string]: any }): IJsonItem[] {
       field: 'deployPort',
       name: t('project.node.mlflow_deployPort'),
       span: deployPortSpan
+    },
+    {
+      type: 'input',
+      field: 'cpuLimit',
+      name: t('project.node.mlflow_cpuLimit'),
+      span: cpuLimitSpan
+    },
+    {
+      type: 'input',
+      field: 'memoryLimit',
+      name: t('project.node.mlflow_memoryLimit'),
+      span: memoryLimitSpan
     }
   ]
 }
@@ -76,5 +98,9 @@ const DEPLOY_TYPE = [
   {
     label: 'DOCKER',
     value: 'DOCKER'
+  },
+  {
+    label: 'DOCKER COMPOSE',
+    value: 'DOCKER COMPOSE'
   }
 ]

dolphinscheduler-ui/src/views/projects/task/components/node/format-data.ts

@@ -355,6 +355,8 @@ export function formatParams(data: INodeData): {
     taskParams.deployModelKey = data.deployModelKey
     taskParams.mlflowProjectRepository = data.mlflowProjectRepository
     taskParams.mlflowProjectVersion = data.mlflowProjectVersion
+    taskParams.cpuLimit = data.cpuLimit
+    taskParams.memoryLimit = data.memoryLimit
   }
 
   if (data.taskType === 'OPENMLDB') {

dolphinscheduler-ui/src/views/projects/task/components/node/tasks/use-mlflow.ts

@@ -49,6 +49,8 @@ export function useMlflow({
     mlflowJobType: 'CustomProject',
     mlflowProjectVersion: 'master',
     automlTool: 'flaml',
+    cpuLimit: '0.5',
+    memoryLimit: '500M',
     mlflowCustomProjectParameters: [],
     delayTime: 0,
     timeout: 30,

dolphinscheduler-ui/src/views/projects/task/components/node/types.ts

@@ -336,6 +336,8 @@ interface ITaskParams {
   deployType?: string
   deployPort?: string
   deployModelKey?: string
+  cpuLimit?: string
+  memoryLimit?: string
   zk?: string
   zkPath?: string
   executeMode?: string