-
-
Notifications
You must be signed in to change notification settings - Fork 523
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
Misbehaviour when inserting 100K docs with float_vectors and HNSW #2883
Labels
Comments
As discussed on today's dev call, this might have something to do with saving data to disk in chunks. I can confirm that. I did further testing. Here’s what it looks like after running the original query:
If you attempt dropping the table after the insert:
The questions are:
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Bug Description:
If you try to insert 100K 512-dims float_vectors, some inserts take too long, and a
DROP TABLE
afterwards also takes too long and doesn't fully remove the table files.Use the manticore-load tool or the cache it generates.
Note, some inserts took up to 29 seconds.
If you attempt dropping the table it takes 49 seconds and can't remove the files:
The same w/o HNSW works fine:
Manticore Search Version:
Manticore 6.3.9 aa1c56713@24122619 dev (columnar 2.3.1 9d9a466@24122408) (secondary 2.3.1 9d9a466@24122408) (knn 2.3.1 9d9a466@24122408)
Operating System Version:
Ubuntu Jammy (dev2)
Have you tried the latest development version?
Yes
Internal Checklist:
To be completed by the assignee. Check off tasks that have been completed or are not applicable.
The text was updated successfully, but these errors were encountered: