"The lesson for us is not every migration has to save money to be worth it; sometimes it is for reduced operational burden or better latency, sometimes it's both. Zerobus Ingest saved us a lot of money, simplified our stack, and put us on a managed path forward. That's rare." - Nick Fornicola, Director of Data & AI Platforms, Beusa Energy
Beusa Energy is using Zerobus Ingest on Databricks to stream telemetry from more than 6,000 devices into its lakehouse with roughly three seconds of end-to-end latency. By replacing a costly SQL-based ingestion path, Beusa simplified its architecture, reduced costs, and created a stronger foundation for near real-time operational insights.
Key highlights:
- More than 99% lower ingestion cost: Replacing the SQL Statement API path reduced ingestion from roughly 689 DBU per GB to 0.29 DBU per GB on the same workload.
- 6,000+ devices streaming at scale: Beusa processes approximately 22 million telemetry rows daily at 1 Hz across more than 250 remote assets.
- Around three seconds to queryable data: Telemetry moves from remote assets into Delta tables with end-to-end latency of about three seconds.
- Simpler, managed architecture: Beusa switched the write path to Zerobus Ingest without changing its MQTT subscriptions, device configurations, or existing worker service, and without adding another streaming broker layer.
- A foundation for predictive maintenance: With high-frequency telemetry alongside maintenance history in the lakehouse, Beusa can continue developing condition-based and predictive maintenance use cases.
🔗 Check out the full story 👈
