About vector search

Preview

This feature is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms. Pre-GA features are available "as is" and might have limited support. For more information, see the launch stage descriptions.

Memorystore for Valkey supports storing and querying vector data. This page provides information about vector search on Memorystore for Valkey.

Important: To use vector search, your instance must be created after the feature launch date of September 13, 2024. If your instance was created prior to this date, you will need to create a new instance to use this feature.

Vector search on Memorystore for Valkey is compatible with the open source LLM framework LangChain. Using vector search with LangChain lets you build solutions for the following use cases:

• Retrieval Augmented Generation (RAG)
• LLM cache
• Recommendation engine
• Semantic search
• Image similarity search

The advantage of using Memorystore to store your Gen AI data, as opposed to other Google Cloud databases is Memorystore's speed. Vector search on Memorystore for Valkey leverages multi-threaded queries, resulting in high query throughput (QPS) at low latency.

Memorystore also provides two distinct search approaches to help you find the right balance between speed and accuracy. The HNSW (Hierarchical Navigable Small World) option delivers fast, approximate results - ideal for large datasets where a close match is sufficient. If you require absolute precision, the 'FLAT' approach produces exact answers, though it may take slightly longer to process.

If you want to optimize your application for the fastest vector data read and write speeds, Memorystore for Valkey is likely the best option for you.