Contribution analytics: Remove request batching when using Clickhouse
Stop restricting request size when using the Clickhouse backend. The batching solution was added to improve load times for PostgresQL, but actually slows the response when fetching from Clickhouse.
Screenshots or screen recordings
There's no visible changes, but you can monitor the GraphQL requests going out to see the change. Instead of requests broken up on a weekly basis, the full time period should be used for the fetched data.
Before | After |
---|---|
![]() |
![]() |
How to set up and validate locally
- Visit the contribution analytics page for any group (ex. http://gdk.test/groups/flightjs/-/contribution_analytics)
- Switch the date range to
Last month
orLast 3 months
- Monitor the GraphQL requests on page load. Each request should be scoped to a single week
Feature.enable(:clickhouse_data_collection)
- Monitor the GraphQL requests again. The requests should no longer be scoped
MR acceptance checklist
This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.
-
I have evaluated the MR acceptance checklist for this MR.
Related to #427601 (closed)
Edited by Alex Pennells