Improvements to query performance for major Jira tables are now live

[#1]

Hello everyone,

Several weeks ago, we announced that we'd make efficiency changes to four major Jira tables in the Atlassian Data Lake schema:

• jira_issue

• jira_issue_cycle_time

• jira_issue_status_history

• jira_project

These changes are now live, so you should see faster performance and dashboard loading times when you query the above tables.

Update on data freshness

In our initial announcement, we mentioned these changes would also cause up to 8-hour delays in data freshness for those tables. Since then, we’ve received concerns about how these changes would impact your use cases for Atlassian Analytics, and we’re very appreciative of the feedback and conversations we’ve had with you.

We’re happy to share that we’ve reduced the processing time on these tables to a 3-hour cycle. Note there will not be impacts on data freshness for the rest of the tables in the schema, which take at least 30 minutes to reflect your product changes.

We’re committed to improving the balance between performance and data freshness. Thank you again for your patience as we continue to investigate and assess further steps we can take.

Leave a comment or contact support if you have any thoughts, questions, or concerns.

[Ben Jackson]

Note for customers using these tables and the 3 hourly cycle time you will see updates in those tables roughly every 3 hours. If you make a data change in Jira that just missed the cycle kickoff then it will be picked up in the next 3 hour cycle. This would see the elapsed time from that change to reflecting in the lake being up to 6 hours.

Based on customer feedback we hope the new 3 hourly cycle will help balance the needs between freshness and performance.

Our early number indicate you should see a significant reduction in query time when using those tables.

We look forward to hearing from you.