You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Mar 27, 2024. It is now read-only.
MongoDB allows a BulkWrite call to be ordered or unordered. Right now all calls to BulkWrite are always ordered. For callers that don't need this, allowing them to specify that the call can be unordered can allow MongoDB to process the operations faster.
The text was updated successfully, but these errors were encountered:
For Amazon DocumentDB, doing a large number of upsert operations in a BulkWrite slow is very slow. They suggested that that we could try breaking up large upsert batch operations into multiple concurrent requests to realize better overall performance. This may be worth trying, but this type of strategy is only possible if we have a way to specify that operations are unordered in the first place.
Original message from DocumentDB support:
if cluster resources are not fully utilized and there are no side effects of concurrent updates, we recommend redistributing your batches across a larger number of concurrent client threads.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
MongoDB allows a BulkWrite call to be ordered or unordered. Right now all calls to BulkWrite are always ordered. For callers that don't need this, allowing them to specify that the call can be unordered can allow MongoDB to process the operations faster.
The text was updated successfully, but these errors were encountered: