Data in Motion — MQTT, Sparkplug B, and Why the Edge Changes Everything
Ignited: Striking Industrial Automation ConversationsShow Notes
OPC UA solves interoperability at the device and system level. But what happens when you need to move data across an entire enterprise — reliably, efficiently, and at scale — without drowning in network traffic or proprietary middleware? That’s where MQTT and Sparkplug B come in. In this episode, Travis and Tom sit down with two legends of the IIoT standards world: Benson Houghland, VP of Products at Opto 22, and Arlen Nipper, co-inventor of MQTT and one of the driving forces behind the Sparkplug specification. Together, they trace the arc from MQTT’s origins as a lightweight publish-subscribe protocol for satellite telemetry to its role as the backbone of modern industrial data infrastructure — and explain why the Sparkplug B specification on top of MQTT gives that data the structure and meaning that enterprise systems need. The conversation gets practical fast: how do decoupled, event-driven architectures compare to traditional polling? What does a real MQTT/Sparkplug deployment look like? And why does edge compute — doing real work close to the source of the data — change the architecture conversation entirely?
More from Ignited: Striking Industrial Automation Conversations
Most manufacturers have more data than they know what to do with, yet none of it speaks the same language or reflects how the factory actually operates. When operational issues arise, teams are often left calling specific individuals who hold critical system relationships entirely in their heads. In this episode, Travis and Tom sit down with Jeff Knepper and Graeme Welton from Flow Software, alongside Ryan Crownover from Vertech, to explore what it takes to build an industrial data architecture that goes beyond basic connectivity and dashboards. They discuss how a true knowledge model maps real-world relationships between assets, processes, people, systems, and events, fundamentally transforming raw operational data into actionable enterprise intelligence.
Traditional industrial historians often struggle to scale across modern, multi-site enterprise environments without introducing support risks or performance bottlenecks. In this episode, we talk with Mike from Tiger Data and Colby from Inductive Automation about their joint partnership designed to deliver an enterprise-grade time-series data backend for Ignition. They share the story behind TimescaleDB’s open-source adoption, the production challenges of running unsupported databases in mission-critical manufacturing, and how their new commercial solution provides high availability, enhanced Hyperfunctions, and scalable deployment options from edge to cloud.
Over 1,900 engineers, integrators, executives, and builders gather at ICC 2026 in Sacramento, California, for what is shaping up to be the biggest Ignition Community Conference ever. To hear what the community is actually thinking, Travis and Tom set up a live studio right at the conference to interview eight community members across various backgrounds. Asking every guest the same set of questions, this episode captures unfiltered insights into why people keep coming back, what caught their attention this year, and what they love and look forward to in the world of Ignition.
Digital transformation initiatives rarely fail because of the software, they fail because the underlying data structure wasn’t built to scale. In this episode, Travis and Tom talk with Dylan DuFresne, Co-Founder and Chief Architect at Abelara, about building industrial architectures designed for long-term viability. Dylan shares a practical, field-tested perspective on the Unified Namespace (UNS) as a core strategy rather than a buzzword, explaining how true data ownership, proper context, and bridging the gap between execution and strategy separate successful manufacturers from those left with disconnected systems.
Connecting plant-floor hardware is a solved problem; the real challenge is making that data act intelligently. In this episode, Travis and Tom sit down with Remus Pop, head of Concept Reply’s industrial AI practice, to discuss his ”Factory Brain” architecture. Remus explains why simply building dashboards isn’t enough, how knowledge graphs bridge the gap between raw data and contextual intelligence, and how emerging tools like MCP and large language models are transforming plant operations. Discover why setting up robust data infrastructure on Ignition today is the key to unlocking AI tomorrow.