andygrove opened a new pull request, #1983: URL: https://github.com/apache/datafusion-ballista/pull/1983
# Which issue does this PR close? None — small, self-contained performance cleanup. # Rationale for this change In `ExecutionGraph::update_task_status`, the task partition id is read via `task_status.clone().partition_id`. This deep-clones the entire `TaskStatus` protobuf message — including its `metrics` vector and nested `status` enum — purely to copy out a single `u32` field, and then immediately drops the clone. `task_status` is already owned (the loop iterates via `into_iter()`) and `partition_id` is `Copy`, so the field can be read directly with no clone. This removes one full `TaskStatus` heap allocation per task-status update on the scheduler's hot path. # What changes are included in this PR? Reads `task_status.partition_id` directly instead of `task_status.clone().partition_id` in the two affected loops in `execution_graph.rs`. No behavior change. # Are there any user-facing changes? No. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
