[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-26。"],[],[],null,["| The [VPC-SC security controls](/vertex-ai/generative-ai/docs/security-controls) and\n| CMEK are supported by Vertex AI RAG Engine. Data residency and AXT security controls aren't\n| supported.\n\nThis page introduces you to `RagManagedDb`, its underlying technology, and how\n`RagManagedDb` is used in Vertex AI RAG Engine. In addition, this page\ndescribes the different tiers that are available to tune performance, which\nmight impact your costs, and provides instructions for deleting your\nVertex AI RAG Engine data, which stops billing.\n\nOverview\n\nVertex AI RAG Engine uses `RagManagedDb`, which is an enterprise-ready,\nfully-managed Google Spanner instance that's used for resource storage\nby Vertex AI RAG Engine and is optionally available to be used as\nthe [vector database of\nchoice](/vertex-ai/generative-ai/docs/rag-engine/use-ragmanageddb-with-rag) for your RAG corpora.\n\nThrough Spanner, Vertex AI RAG Engine offers a\nconsistent, highly available, and highly scalable database to support your\napplication. To learn more about Google Spanner, see\n[Spanner](/spanner).\n\nVertex AI RAG Engine stores your RAG corpus and RAG file resource\nmetadata in `RagManagedDb`, regardless of your choice of vector database. Vector\ndatabases are only used for storage and retrieval of embeddings. In addition to\nresource storage, `RagManagedDb` can also be used to store and manage vector\nrepresentations of your documents. The vector database is then used to retrieve\nrelevant documents based on the document's semantic similarity to a given query.\n\nManage tiers\n\nVertex AI RAG Engine lets you scale your `RagManagedDb` instance based\non your usage and performance requirements using a choice of two tiers, and\noptionally, lets you delete your Vertex AI RAG Engine data using\na third tier.\n\nThe tier is a project-level setting that's available in the `RagEngineConfig`\nresource that impacts RAG corpora using `RagManagedDb`. The following tiers\nare available in `RagEngineConfig`:\n\n- **Scaled tier**: This tier offers production-scale performance along with\n autoscaling functionality. It's suitable for customers with large amounts of\n data or performance-sensitive workloads. Internally, this tier sets the\n Spanner instance to autoscaling configuration with a minimum\n of 1 node (1,000 processing units) and a maximum of 10 nodes (10,000\n processing units).\n\n- **Basic tier (default)**: This tier offers a cost-effective and low-compute\n tier, which might be suitable for some of the following cases:\n\n - Experimenting with `RagManagedDb`.\n - Small data size.\n - Latency-insensitive workload.\n - Use Vertex AI RAG Engine with only other vector databases.\n\n To offer the Basic tier, `RagManagedDb` sets the underlying\n Spanner instance to a fixed configuration of 100 processing\n units, which is equivalent to 0.1 nodes.\n- **Unprovisioned tier** : This tier deletes the `RagManagedDb` and its\n underlying Spanner instance. The Unprovisioned tier disables\n the Vertex AI RAG Engine service and deletes your data held\n within this service regardless of the vector database used for your\n `RagCorpora`. This stops the billing of the service. For more information on\n billing, see [Vertex AI RAG Engine\n billing](/vertex-ai/generative-ai/docs/rag-engine/rag_engine_billing).\n\n After the data is deleted, the data can't be recovered. To start usingVertex AI RAG Engine again, you must update the tier by\n calling the `UpdateRagEngineConfig` API.\n\n| **Note:** The Enterprise tier from the `v1beta1` version was renamed to the Scaled tier.\n\nGet the project configuration\n\nThe following code samples demonstrate how to use the `GetRagEngineConfig` API\nfor each type of tier:\n\n- [Version 1\n (v1)](/vertex-ai/generative-ai/docs/model-reference/rag-api-v1#get_project_configuration) API\n code samples.\n\n- [v1beta1](/vertex-ai/generative-ai/docs/model-reference/rag-api#get-project-config-for-rag) API\n code samples.\n\nUpdate the project configuration\n\nThe following code samples demonstrate how to use the `UpdateRagEngineConfig`\nAPI for each type of tier:\n\n- [Version 1\n (v1)](/vertex-ai/generative-ai/docs/model-reference/rag-api-v1#update_project_configuration)\n API code samples.\n\n- [v1beta1](/vertex-ai/generative-ai/docs/model-reference/rag-api#update-project-config-for-rag)\n API code samples.\n\nWhat's next\n\n- To learn how to use the RAG API v1, the default, see [RAG API\n v1](/vertex-ai/generative-ai/docs/model-reference/rag-api-v1).\n- To learn how to use the RAG API v1beta1, see [RAG API\n v1beta1](/vertex-ai/generative-ai/docs/model-reference/rag-api).\n- To learn more about `RagManagedDb` and how to manage your tier configuration as well as the RAG corpus-level retrieval strategy, see [Use RagManagedDb with\n Vertex AI RAG Engine](/vertex-ai/generative-ai/docs/rag-engine/use-ragmanageddb-with-rag)."]]