Azure AI Search Connector > Introduction to Azure AI Search Connector
  

Introduction to Azure AI Search Connector

You can use Azure AI Search Connector to write unstructured text files, such as PDFs, as vector embeddings to the Azure AI Search vector database.
You can use an Azure AI Search connection as a target in mappings in advanced mode. When you run a mapping in advanced mode or a task based on a mapping in advanced mode, Data Integration analyzes the data logic in the mapping to automatically assign data logic to run on an advanced cluster.

Example

You run a legal firm that handles a massive repository of legal documents, contracts, and case files in PDF format in Microsoft Azure Data Lake Storage Gen2.
Legal professionals need efficient ways to search and retrieve relevant documents quickly, often using complex queries that go beyond simple keyword searches.
To enhance the speed and accuracy of document retrieval, you want to integrate your data into the Azure AI Search vector database. This integration allows legal professionals to efficiently query high-dimensional vectors that represent complex data, such as text and tables, streamlining their search processes and enhancing overall productivity.
To integrate data in Azure AI Search, you perform the following steps:
  1. 1Create a mapping in advanced mode.
  2. 2Add a Source transformation to read data from Microsoft Azure Data Lake Storage Gen2.
  3. 3Add a Chunking transformation to split the text into smaller manageable chunks for enhanced vector representation.
  4. 4Add an Embedding transformation to convert the text to vector embeddings.
  5. 5Add a Target transformation to use the Azure AI Search connection and load the vector embeddings to Azure AI Search.
  6. 6Run the mapping in advanced mode.
Preview Notice: Azure AI Search Connector is available for preview. Preview functionality is supported for evaluation purposes but is unwarranted and is not supported in production environments or any environment that you plan to push to production. Informatica intends to include the preview functionality in an upcoming release for production use, but might choose not to in accordance with changing market or technical circumstances. Note that if you are working on a preview POD, all data is excluded from SOC 2 compliance coverage. For more information, contact Informatica Global Customer Support.