Alibaba Cloud Elasticsearch AI Engine Edition is a cloud-native search engine for retrieval-augmented generation (RAG), code search, agent memory, and multi-tenant search.
AI Engine Edition is built on the Elasticsearch 9 series. It decouples persistent storage from indexing and query workloads: index data and persistent state reside in object storage, index nodes handle writes and index building, and search nodes handle queries. Applications connect through a Collection as a unified access point and organize knowledge bases, code repositories, agents, or tenants as Slices, without creating a separate index for each one.
Topics
The following table describes the topics in this section.
| Topic | Description |
| What is AI Engine Edition | Product positioning, core advantages and applicable scenarios, Collection and Slice concepts, read-write and query architecture, DiskBBQ disk-native vector index, performance reference, and version and compatibility. |
| AI Engine Edition usage guide | Instance creation and connection, Collection mode selection, and quick-start paths for namespace mode and vector cluster recall mode; Collection configuration, Slice management, data write, read, and query operations, Slice cache warmup, DiskBBQ configuration, Collection Alias, and Copy Slice; Collection observability, permissions, feature boundaries, and API reference. |
Next steps
If you are new to AI Engine Edition, complete the following steps:
Read What is AI Engine Edition to confirm the product capabilities and applicable scenarios.
Follow AI Engine Edition usage guide to create an instance, ingest data, and verify queries.