Viewing partition recommendations

You can view and analyze the collected data, which includes recommendations for microservices to implement in place of the Java® monolithic application.

About this task

The AI engine for partition recommendations generates partition recommendations along with runtime invocation details by analyzing the runtime traces of the completed use cases for the Java® monolithic application. These partitions are groupings of monolith Java® classes that serve as starting points for potential microservices. The AI engine also generates class containment dependency details from the static code analysis of the Java® monolithic application.

The generated partition recommendations, the class dependency analysis, and the runtime invocation analysis of a Java® monolithic application are contained in the final_graph.json file in the mono2micro-output/oriole directory. Use the IBM® Mono2Micro™ workbench UI by running the mono2micro workbench command to view and customize these partition recommendations.

Prerequisites

Docker or Podman is required to run the workbench UI.

Procedure

  1. Install Docker or Podman in your workstation or in a local computer where you plan to run the mono2micro workbench command.
  2. Use the mono2micro workbench command to start the workbench UI.
  3. Load the JSON file from the mono2micro recommend or mono2micro refine command output to the workbench UI.
  4. View the partition recommendations in the graph and table views.
  5. Create a custom view where you can modify the recommendations to fit your needs.
  6. Save your partitions so you can feed them back into the analyzer to take the next steps.

What to do next

You can now proceed to Accessing the workbench.

For information about terms in partition recommendations, see Concepts and terminology.