Learn extra at:
You should utilize the identical question instruments to look vector indexes in addition to the remainder of your information, providing you with the choice to look primarily based on similarities in your information or by actual matches. This method is just like how large-scale search engines like google work and can assist discover and rank outcomes from massive semistructured information units, for instance, trying to find related critiques on an e-commerce web site. Material requires a vector coverage for every Cosmos DB container, which defines measurement, dimensionality, and the underlying distance perform used to seek for comparable vectors. Search applied sciences like DiskANN require a excessive dimensionality, with a minimum of 1,000 dimensions (and a most of 4,096).
Querying Cosmos DB in Material
If you query data stored in Cosmos DB through Fabric’s OneLake, you’re working with a mirrored copy of your Cosmos DB data. As you retailer information, it’s copied throughout within the Delta Parquet format utilized in Material, permitting you to make use of any of the supported question instruments, together with the desktop Energy BI for advert hoc evaluation. Queries right here may be made throughout all of your operational information, not simply Cosmos DB, treating it as a unified entire and nonetheless making the most of Cosmos DB’s characteristic set for functions that want to make use of that information.
This additionally means that you can reap the benefits of different Material options together with your Cosmos DB information, for instance, utilizing it to rapidly add embeddings and a vector index to your information, so it may be used as a part of the grounding information for an AI utility primarily based on retrieval-augmented generation (RAG).