Database write overhead reduced for task tags
The system no longer creates redundant join table records for task tags, reducing database write overhead and simplifying data retrieval.
The system previously saved task tags in two places: a simple array and a complex relational table, which created redundant database writes during high-volume task triggering. The relational tag tables are now completely bypassed in favor of a single denormalized array.
This drastically reduces database load during task execution, stripping out unnecessary cross-table associations and simplifying data retrieval. The improvements touch the API, the web app, and the core run engine, removing friction across the stack.
View Original GitHub Description
The TaskRun.runTags string array already stores tag names, making the TaskRunTag M2M relation redundant write overhead. Remove createTags calls, connect: tags, and join table writes from both V1 and V2 trigger paths. Simplify the add-tags API to just push to runTags directly.