UNRWA Digital archive
Services

Azure Blob Storage

How Azure Blob Storage is used across the pipeline to stage, classify, and persist refugee documents.

Overview

Azure Blob Storage is used throughout the pipeline as the primary document store. It replaces SharePoint as the working source for all processing stages, providing a more reliable and scalable foundation for document handling.

Containers

ContainerPurpose
unprocessed-filesReceives documents uploaded from SharePoint during ingestion. These are raw, unclassified files awaiting processing.
master-cardsStores classified Master Card documents.
index-cards (Not yet implemented)Stores classified Index Card documents.
red-cross-cardsStores classified Red Cross Card documents.
unclassified-filesStores documents that could not be classified — damaged, unreadable, or unknown format. These require manual review.

Role in the Pipeline

  • Ingestion — Power Automate uploads documents from SharePoint into unprocessed-files
  • Classification — the Document Classification service reads from unprocessed-files and routes each document to its type-specific container
  • Data Extraction — the Data Extraction service fetches documents from type-specific containers and sends them to Azure Document Intelligence

On this page