-
Notifications
You must be signed in to change notification settings - Fork 5
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
Scalar quantization reliability test #228
base: main
Are you sure you want to change the base?
Conversation
apps/replicated-import/run.py
Outdated
"vectorIndexConfig": { | ||
"sq": {"enabled": True, "trainingLimit": 10000, "rescoreLimit": 20}, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @donomii, this should fail due to additional validation weaviate/weaviate#5341. Both PQ and SQ require that either async indexing is enabled or that they are enabled post loading some data. You can use this import script on the other PR to show how importing is paused while PQ/SQ is enabled
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I added the ASYNC_INDEXING environment variable, since this seemed to make the most sense for the test. So now the test confirms that a node can crash while objects are being queued for later indexing, and import still works.
Update schema to use bigram vectors and scalar quantization Use bigram vectorizer in trigram mode to give (sparse) 16k vectors Print container name ? Add the actual shell script linter Actually run the test
3166308
to
260d734
Compare
This reverts commit f65942b.
…te/weaviate-chaos-engineering into scalar-quantization-reliability-2
Adds a test that imports long sparse vectors to a cluster in Scalar Quantization mode, and randomly crashes a node to simulate cluster failure.