This topic describes the architecture of Pivotal Scheduler.

HA Topology

In a highly available resource configuration, the Pivotal Scheduler service uses the following:

  • Three Scheduler instances, each running the Scheduler HTTP API server and a Scheduler Engine.

  • Three Scheduler Service Broker instances

You can reduce the number of Scheduler instances to one using the Cloud Foundry Command Line Interface (cf CLI).

Data Persistence

Scheduler relies on a MySQL datastore to persist data, including jobs, calls, and history. Each call or schedule can require up to 10 MB of database capacity to store history.