In this code, using a python Jupyter Notebook we ingest a large number of documents from a storage account (or you can use a SharePoint Site, code is there), save the chunked information into an Azure SQL database using a stored procedure or Azure AI Search.
The stored procedure saves data to the documents table - or Azure AI Search, saves the embeddings and creates similarity vector table, as well as saving key phrases into a graph table for searching.