TheVPC-SC security controls and CMEK are supported by Vertex AI RAG Engine. Data residency and AXT security controls aren't supported.

This page introduces you toRagManagedDb, its underlying technology, and howRagManagedDb is used in Vertex AI RAG Engine. In addition, this pagedescribes the different tiers that are available to tune performance, whichmight impact your costs, and provides instructions for deleting yourVertex AI RAG Engine data, which stops billing.

Vertex AI RAG Engine usesRagManagedDb, which is an enterprise-ready,fully-managed Google Spanner instance that's used for resource storageby Vertex AI RAG Engine and is optionally available to be used asthevector database ofchoice for your RAG corpora.

Through Spanner, Vertex AI RAG Engine offers aconsistent, highly available, and highly scalable database to support yourapplication. To learn more about Google Spanner, seeSpanner.

Vertex AI RAG Engine stores your RAG corpus and RAG file resourcemetadata inRagManagedDb, regardless of your choice of vector database. Vectordatabases are only used for storage and retrieval of embeddings. In addition toresource storage,RagManagedDb can also be used to store and manage vectorrepresentations of your documents. The vector database is then used to retrieverelevant documents based on the document's semantic similarity to a given query.

Manage tiers

Vertex AI RAG Engine lets you scale yourRagManagedDb instance basedon your usage and performance requirements using a choice of two tiers, andoptionally, lets you delete your Vertex AI RAG Engine data usinga third tier.

The tier is a project-level setting that's available in theRagEngineConfigresource that impacts RAG corpora usingRagManagedDb. The following tiersare available inRagEngineConfig:

• Scaled tier: This tier offers production-scale performance along withautoscaling functionality. It's suitable for customers with large amounts ofdata or performance-sensitive workloads. Internally, this tier sets theSpanner instance to autoscaling configuration with a minimumof 1 node (1,000 processing units) and a maximum of 10 nodes (10,000processing units).

• Basic tier (default): This tier offers a cost-effective and low-computetier, which might be suitable for some of the following cases:

  • Experimenting withRagManagedDb.
  • Small data size.
  • Latency-insensitive workload.
  • Use Vertex AI RAG Engine with only other vector databases.

  To offer the Basic tier,RagManagedDb sets the underlyingSpanner instance to a fixed configuration of 100 processingunits, which is equivalent to 0.1 nodes.

• Unprovisioned tier: This tier deletes theRagManagedDb and itsunderlying Spanner instance. The Unprovisioned tier disablesthe Vertex AI RAG Engine service and deletes your data heldwithin this service regardless of the vector database used for yourRagCorpora. This stops the billing of the service. For more information onbilling, seeVertex AI RAG Enginebilling.

  After the data is deleted, the data can't be recovered. To start usingVertex AI RAG Engine again, you must update the tier bycalling theUpdateRagEngineConfig API.

• Version 1(v1) APIcode samples.

• v1beta1 APIcode samples.

Update the project configuration

The following code samples demonstrate how to use theUpdateRagEngineConfigAPI for each type of tier:

• Version 1(v1)API code samples.

• v1beta1API code samples.

What's next

• To learn how to use the RAG API v1, the default, seeRAG APIv1.
• To learn how to use the RAG API v1beta1, seeRAG APIv1beta1.
• To learn more aboutRagManagedDb and how to manage your tier configurationas well as the RAG corpus-level retrieval strategy, seeUse RagManagedDb withVertex AI RAG Engine.