Case Study: BreezoMeter achieves real-time, hyper-local air quality at global scale and 70% infrastructure savings with Google Cloud Platform

BreezoMeter Mapping the world’s air with Google Cloud Platform

BreezoMeter delivers real-time, hyper-local air quality data via an API, covering millions of people across dozens of countries. The company must ingest and validate thousands of disparate sensor sources and run hourly models that generate and process roughly 1.5 TB of data per calculation to estimate 17 pollutants at hundreds of millions of locations — a massive, time-sensitive compute and scaling challenge.

BreezoMeter migrated its processing and API to Google Cloud Platform, using App Engine for high-availability request handling, Compute Engine custom machines and Preemptible VMs for heavy modeling (cutting infrastructure costs by 70%), Nearline for automated archiving, and Kubernetes, Pub/Sub and Dataflow for scalable microservices and ingestion. The result: reliable hourly calculations and millions of API requests served with automated cost-efficient storage and rapid expansion — adding 35 countries in two weeks and enabling plans to scale coverage to billions more.

BreezoMeter

Emil Fisher

Co-Founder & CTO