The Data Center Frontier Show Cooling as a Service: Rethinking the Economics of AI Infrastructure
6 snips
Jan 29, 2026 Mike Obradovich, VP of Global Accounts for Data Centers at Ecolab, provides expertise in cooling solutions and Cooling as a Service. He discusses CaaS as a managed performance layer with predictable SLAs. He highlights proactive, analytics-driven maintenance and 3D telemetry for trend-based alerts. He covers co-design with providers and treating cooling as an integrated ecosystem for AI workloads.
AI Snips
Chapters
Transcript
Episode notes
Cooling Treated As A Managed Performance Layer
- Cooling as a Service reframes cooling from a fixed asset to a managed performance layer with embedded expertise.
- This approach shortens decision cycles, reduces commissioning rework, and brings predictable performance SLAs for clearer CapEx/OpEx planning.
Shift From Reactive To Predictive Maintenance
- Move maintenance from reactive to predictive by pairing remote analytics with on-site rapid response.
- Use continuous fluid health telemetry to catch chemistry drifts that cause corrosion, fouling, and microbial growth.
Telemetry Enables Early Intervention
- Real-time telemetry like 3D Tracer provides leading indicators rather than lagging lab data.
- Aggregating telemetry and lab analytics converts trends into targeted, scheduled interventions that extend fluid life.