[SOLR-15880] Introduce Support to K Nearest Neighbors Search

• Type: New Feature
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 9.0
• Fix Version/s: 9.0
• Labels: vector-based-search
• Assignee: Alessandro Benedetti
• Reporter: Alessandro Benedetti
• Created: 24/Dec/21 10:24
• Updated: 29/Feb/24 05:53
• Resolved: 28/Jan/22 03:47

Description

This contribution introduces in Apache Solr the ability to run vector-based searches using native Apache Lucene data structures optimized for approximate nearest neighbours.
This is the first milestone in introducing Neural Search in Apache Solr.
More will come later.

It builds on top of https://issues.apache.org/jira/browse/LUCENE-9004 to provide a:

• Dense Vector Field - At moment Single valued, supporting stored, indexed
• Knn Query Parser - simple query parser to run knn search on target vector and field

The Dense Vector field only supports KNN searches (no range query, no field queries, no existence queries...)

The Knn query parser can be used as query, filter query and rerank query.

Issue Links

• causes: SOLR-16502 Multiple CopyField should not limit to first maxChars
• depends upon: LUCENE-9004 Approximate nearest vector search
• is related to: SOLR-14397 Vector Search in Solr
• supercedes: SOLR-14397 Vector Search in Solr
• links to: GitHub Pull Request #476