Connection object in Azure AI Search pointing an indexer at its content, which must be a supported source: Blob, ADLS Gen2, Table Storage, Cosmos DB, Azure SQL, SQL Managed Instance or SQL Server on Azure VMs. One indexer, one data source.
Also called data source.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Search data source in context, with comparison tables and the common traps.
Terms in this definition
- Connection object
Usually generated automatically by the KCC, this object tells a DC which partner it pulls replication changes from (inbound, one direction only).
- Azure AI Search
Managed Azure service that builds indexes over your content and answers keyword, vector, hybrid and semantically ranked queries, optionally with AI enrichment. Retrieval-augmented generation, Foundry agents and knowledge mining all use it to fetch relevant content.
- Indexer
The pull-model crawler in Azure AI Search that takes data from a single data source, can apply a skillset and loads the output into a single index. It runs on demand or on a schedule, at most every five minutes.
- BLOB
Short for binary large object: raw binary data, like pictures, audio, video or documents, that only an application can make sense of. Azure keeps these files as blobs in Blob Storage.
- ADLS Gen2
Short for Azure Data Lake Storage Gen2: a standard GPv2 account with hierarchical namespace turned on, so analytics workloads get true directories and POSIX-style ACLs.
- Table storage
Low-cost key-value tables indexed solely on RowKey and PartitionKey, limited to a single write region and entities of 1 MB.
- Cosmos DB
NoSQL database distributed globally, offering writes in multiple regions, automatic indexing and latency below 10 ms.
- Auditing
Azure SQL capability that sends database audit logs to Log Analytics, Event Hubs or a storage account; that account is allowed to be in a different region.
Related terms
- Calculated table
Instead of coming from a data source, this kind of model table is defined by a DAX formula. It's always held in Import mode and recomputed on each refresh, which makes it handy for staging intermediate results or duplicating a table for role-playing.
- Parallel indexing
Runs several indexers at once, one per partition of the content (a container or virtual folder), each with a dedicated data source but sharing one skillset and target index. Concurrency is capped at about one indexer job per search unit.
- Power BI template
A .pbit file containing pages, model definition and queries from a report, with no data, to start new reports from. When someone opens it, they are asked for parameter values and data source sign-in details.
- Security Copilot plugin
Lets Security Copilot reach a data source, whether Microsoft, third-party or custom. Users still need their own permissions on that source for it to return anything.
- Take over
Only an owner can set up scheduled refresh, data source credentials and automatic aggregations for a semantic model. This action, in the model's settings in the service, transfers ownership to whoever selects it.
- View Native Query
A Power Query command, also labelled View data source query, that displays what a step sends to the source. If it is greyed out on a step where the source normally supports it, query folding has stopped at that step.