Optimization Server Master is a REST-API application that stands for the main entry point to the Optimization server.

The Master application allows the clients to create jobs and follow their execution.

The following features are available through the REST API :

  • List and get details on the available tasks.
  • CRUD operations on jobs.
  • Start and stop jobs.
  • Access the job input and output (the buckets). Buckets have an identifier and a byte array content.

For more information, you can read the full API documentation in the Master API Sub-section.

Tasks, jobs, buckets and job events are stored in MongoDB by the master, to be accessible at any time by a client. Buckets under 15MB are stored in the standard MongoDB way. Due to a MongoDB limitation, buckets over 15MB are stored using GridFS. All data have a Time To Live (one week by default), after which they are deleted. TTL can be overridden using a spring property described in Deployment Section.

Master Limits

The Master application also allows to limit the number of running jobs per label.

For more information, you can see the detailed documentation in the Master Configuration Sub-section