Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Mas d34 leveled.i465 stopfold #467

Merged
merged 3 commits into from
Jan 15, 2025

Conversation

martinsumner
Copy link
Owner

@martinsumner martinsumner commented Jan 10, 2025

Improve efficiency of fold_objects sqn_order - and hence efficiency of handoffs in Riak

#465

(also includes - #462)

The riak_SUITE:handoff_close/1 test would previously fail, as there would be around 5s of time processing changes after the fold to the JournalSQN has finished. It now tests that this is less than 1s (on a fats machine should be < 10ms).

The batch size has also bene reduced to alternate more frequently between pushing disk and pushing CPU.

Test previously fails, as even on a fast machine the fold goes on for 5s beyond the last object found.

With change to reduce batch size, and stop when batch goes beyond JournalSQN - success with << 100ms spent folding after the last object discovered
As the key changes are going to be discarded
@martinsumner martinsumner merged commit 03a2092 into develop-3.4 Jan 15, 2025
2 checks passed
martinsumner added a commit to OpenRiak/leveled that referenced this pull request Jan 15, 2025
* Test and fix - issue with folding beyond JournalSQN

Test previously fails, as even on a fast machine the fold goes on for 5s beyond the last object found.

With change to reduce batch size, and stop when batch goes beyond JournalSQN - success with << 100ms spent folding after the last object discovered

* Wait after suite for delete_pending to close

martinsumner#462

* Avoid processing key changes in object fold runner

As the key changes are going to be discarded
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants