Case Studies Healthcare
Global medical device leader
Device telemetry — hub & spoke
Device telemetry over MQTT gateways in a regional hub-and-spoke design—real-time signals and historicals on one backbone; AsyncAPI for REST adoption. 500K+ devices globally; 30B+ events/day.
Apache Kafka · Apache Flink · MQTT · Confluent · Databricks · Apache Spark · AsyncAPI
-
Customer
A global medical device leader instrumenting fleets in the field. Telemetry feeds product, operations, and analytics—not a side channel.
-
Challenge
Device traffic arrived through MQTT gateways across regions. Real-time signals and historical retention had to share one trustworthy backbone. Copy-everything-to-the-lake failed on both speed and control.
Interface sprawl made producer and consumer contracts hard to adopt. Teams needed a governed way to expose and consume the backbone, including REST-shaped access.
-
Work
Owned the data backbone for device telemetry: MQTT gateways feeding a regional hub-and-spoke architecture for both real-time signals and historicals.
Drove REST interface adoption via AsyncAPI so contracts were explicit. Kafka, Flink, and Confluent for movement; Databricks and Spark on the lakehouse path so analytics stayed aligned to the same events.
-
Impact
- 500K+ devices managed globally on the telemetry backbone
- 30B+ events/day through the hub-and-spoke design
- Real-time signals and historicals on one backbone; AsyncAPI-governed REST adoption
