Skip to content

FHIR Files Connector

The FHIR Files Connector ingests healthcare data from flat files into the FHIR service deployed within your Sidra environment. Leveraging Sidra’s connector framework, it delivers fast, reliable loading of large NDJSON datasets stored in Azure Blob Storage.

Each NDJSON file is newline-delimited, with every line representing a single FHIR resource—such as a Patient, Condition, or Observation. The connector downloads these files in configurable parallel batches, parses them, and upserts the resulting resources to Sidra’s FHIR server.

Prerequisite
The Sidra FHIR service must be installed and reachable at run time. The connector does not perform an automatic availability check.

Key features

  • Parallel batch processing for high throughput
  • Incremental loading — each run starts with files newer than the last successfully processed one
  • Delta tracking — records the name and timestamp of every ingested file in a Delta table

How it works

  1. Discover files in the configured storage container, ordered by last-modified date
  2. Skip processed files by consulting the Delta tracking table
  3. Download and parse files concurrently, respecting the chosen batch size
  4. Upsert resources to Sidra’s FHIR server via its REST API
  5. Audit success by logging each processed filename in the Delta table

When to use it

  • Migrating NDJSON exports into Sidra’s governed FHIR landing zone
  • Scheduled ingestion of clinical data files delivered by external partners
  • Feeding downstream Data Products that rely on standardized FHIR data