Operations | Monitoring | ITSM | DevOps | Cloud

In this episode of FinOps on Azure, Michael Stephenson sits down with Frank Contrepois, independent FinOps voice and co-host of The FinOps Guys podcast — to explore what it really means to manage cloud costs from a business-first perspective. Frank has been in the FinOps space for nearly a decade and brings a genuinely different angle to the conversation. His background in commodity trading at Strategic Blue (a Morgan Stanley spinoff) shaped how he thinks about reserved instances, commitment strategies, and why most teams approach cost management the wrong way round.

Understand the real difference between observability and monitoring — and why modern IT teams in 2026 need both. Monitoring tells you something is broken; observability explains why. See real examples, faster troubleshooting workflows, and how Motadata ObserveOps unifies both in one platform. Don’t forget to like, share, and subscribe for more IT insights.

In this episode of Masters of Data, we get into the messier side of AI adoption, tackling questions like who actually owns the output when AI gets it wrong, and whether chasing efficiency is making us forget what it means to be human in the first place. We discuss tech CEOs proudly announcing they no longer think for themselves and debate whether AI is quietly eroding our critical thinking skills. We make the case that purpose-built, narrow AI is genuinely exciting, but that no efficiency gain is worth losing the human touch that makes work, connection, and creativity meaningful.

Cheryl Nomanson, SolarWinds staff technical trainer, provides a comprehensive overview of SolarWinds architecture and scaling options for self-hosted deployments. She explains the centralized deployment model starting with a single SolarWinds server that handles polling, web console, and database connections. The presentation covers key scaling indicators including polling thresholds that warn users at 85% capacity and alert at 100%. She demonstrates how to add up to 100 polling engines per server and additional web servers to handle more concurrent users.