[CASSANDRA-18557] CEP-30 ANN Vector Search with SAI

• Type: Epic
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 5.0-alpha2, 5.0
• Component/s: Feature/Vector Search
• Epic Name: CEP-30 ANN Vector Search with SAI
• Change Category: Operability
• Complexity: Byzantine
• Platform: All
• Assignee: Mike Adamson
• Reporter: Jonathan Ellis

Authors

Mike Adamson

• Created: 31/May/23 02:56
• Updated: 02/Jan/24 12:01
• Resolved: 21/Dec/23 18:04

Description

CEP-30

1. Implement approximate nearest neighbor (ANN) vector search capability in Apache Cassandra using storage-attached indexes (SAI).
2. Support a vector of float32 embeddings as a new CQL type.
3. Add ANN search to work with normal Cassandra data flow (insertion, updating, and deleting rows). The implementation should support adding a new vector in log(N) time, and ANN queries in M log(N) time where N is the number of vectors and M is the number of sstables.
4. Compose with other SAI predicates.
5. Scatter/gather across replicas, combining topK from each to get global topK.

Issue Links

• is related to: CASSANDRA-18604 Write docs for CEP-30 Vector Search