Jira MAR Management

NOTE: We wrote this article before we changed our pricing model on February 1, 2022. If you signed up for a new Fivetran account or renewed your existing account after January 31, 2022, you have an unlimited number of free customer-triggered re-syncs per connector each month. For more information about MAR usage, see our pricing documentation.

Use case

You want to understand how the functioning of Jira connectors affects Monthly Active Rows (MAR).

Environment

Jira Connectors

Jira connector overview

The Jira connector uses the Jira platform REST API and Jira software REST API endpoints. The schema output from the Jira connector is predefined, but you can select or deselect some table groups from syncing. Jira’s own ERD is very complex and Fivetran simplifies these objects into more analytics-friendly relationships. Depending on when your Jira connector was created, we organize some tables differently. For connectors created before September 10, 2020, we created an ISSUE_<FIELD_NAME> table per each array field and an ISSUE_<FIELD_NAME>_HISTORY table per each field for storing historical changes. After September 10, 2020, we only create the ISSUE_FIELD_HISTORY table to store historical changes for non-array fields as well as current values of custom non-array fields, and the ISSUE_MULTISELECT_HISTORY table to store historical changes for array fields as well as current values of custom array fields.

Contributing factors

There are the following primary factors that contribute toward MAR:

• Number of issues created
• Number of times an issue field gets updated or created
• Number of users performing actions in Jira (for example, the COMMENTS table)
• The main tables that contribute to MAR, depending on when the connector was created:
  • After September 10, 2020, the ISSUE_FIELD_HISTORY and ISSUE_MULTISELECT_HISTORY tables
  • Before September 10, 2020, the ISSUE_<FIELD_NAME> and ISSUE_<FIELD_NAME>_HISTORY tables

Managing MAR

• There is not much you can do to reduce the MAR from this connector since you can only deselect some table groups from syncing. You can create a Jira user that does not have access to specific objects (for example, boards, etc.) in Jira, and Fivetran then skips these when syncing data to the destination.
• The ISSUE_FIELD_HISTORY and ISSUE_MULTISELECT_HISTORY, or ISSUE_<FIELD_NAME> and ISSUE_<FIELD_NAME>_HISTORY tables, depending on the date the connector was created, are the main tables contributing to MAR.
• You can potentially fully sync some table groups.