EDMM Framework

EDMM provides a declarative model describing the components to be deployed, their configurations, required artifacts, and relations among them. The resulting EDMM model is independent of any specific deployment technology and can be exported from an EDMM-enabled modeling tool or created directly using a text editor according to the respective YAML specification. This model can be fed into the EDMM Transformation and Deployment Framework, which provides a command-line interface (CLI) that can either be used directly by the user or integrated into any automation workflow. The CLI can be used to select the desired target deployment technology into which the EDMM model is to be transformed. The output is an executable technology-specific deployment model that can be executed with the selected technology. The following figure shows an high-level overview of the EDMM Transformation Framework.

DEMO: https://youtu.be/MGWBFomOGYU

The EDMM Modeling Tool, which is based on Eclipse Winery, in order to graphically model an EDMM-based application deployment. From here, users can export an EDMM YAML file according to the specification or trigger the transformation and download an archive containing the generated files and templates for a certain deployment technology. Similarly, the EDMM CLI can be used to trigger the same transformation from a command-prompt:

edmm transform kubernetes <path-to-edmm-yaml-file>

The generated files are usually located relative to the given model file. Finally, users can employ a deployment technology's native tooling to execute the deployment of the application.

Usage: EDMM CLI

The final distribution package can be downloaded form the releases page. Extract the files to a location on your filesystem and add it to your path variable (Linux: $PATH, Windows: %PATH%). Afterwards you can invoke the edmm command from a command prompt.

The transformation framework supports YAML files as input, according to the published YAML specification. Specified components, and their respective component types, must be supplied in a single file at the moment. However, we introduce a couple of built-in component types that can be used to model an application deployment. An example shows the usage of the built-in types to model an application deployment based on virtual compute resources, e.g., virtual machines having some software components installed.

The transformation can be started by using the transform command of the edmm tool: edmm transform <target> <input>. You have to specify the target technology (e.g., one of "ansible", "azure", "chef", "compose", "heat", "kubernetes", "terraform") and the input EDMM YAML model file. The generated technology-specific deployment models will be stored relative to the YAML input file.

Plugins

Each plugin implements its own transformation logic by providing a respective Plugin implementation. Further, a plugin implements different lifecycle methods, e.g., checkModel(), prepare(), and cleanup(), but must provide an implementation for the transform() method. Currently we support the following list of plugins:

• Ansible
• Azure Resource Manager
• Chef
• Docker Compose
• Heat Orchestration Template
• Kubernetes
• Terraform
• Puppet
• Cloudify
• AWS CloudFormation
• Salt
• Juju
• CFEngine

The corresponding plugin README.md contains detailed information of the transformation rules each plugin employs.

Built-in Types

The framework provides built-in types that can be used to model in EDMM. Simply copy and paste the types to your model file. An example showing the usage of these types is available here.

Build the project

We use Maven as our build tool:

./mvnw clean package

Build the Docker image:

docker build -t edmm .
docker run -it -p 5000:5000 edmm