HiveMQ on Scaling MQTT & the Unified Namespace - Ep. 047

Recorded live at Inductive Automation’s ICC 2025 in Sacramento, we sit down with Magnus McHune (HiveMQ) to unpack what it really takes to scale MQTT, build a unified namespace (UNS) that sticks, and bridge the OT/IT divide without blowing up cost or complexity. We dig into HiveMQ’s roots in connected vehicles, why manufacturers are standardizing on MQTT + UNS, what “leveling up” looks like with Ignition 8.3 and new Ignition module work, and how to think about cost, complexity, and cybersecurity as your data volume 10× or 100×. We also touch on HiveMQ Pulse (context + ontology for industrial data) and practical start-up patterns (K8s/VMs, clustering, QoS, topic design).

You’ll learn:

• How to design MQTT topics/payloads that scale

• When open source hits limits—and where enterprise support fits

• Architectures for edge ↔ site ↔ central clusters

• Why outbound-only MQTT + TLS simplifies security models

• How context/semantics (Pulse) unlock AI/ML value

Presented by OpSite Energy — Insight-Driven Remote Optimization.

Links & resources

- OpSite Energy: https://www.opsiteenergy.com

- Energy Tech Podcast: https://www.opsiteenergy.com/all_podcast

- HiveMQ (MQTT platform): https://www.hivemq.com

- Ignition by Inductive Automation: https://inductiveautomation.com

- Learn about HiveMQ Pulse: https://www.hivemq.com/solutions/

00:00 Intro (Energy Tech Podcast @ ICC 2025)

00:35 Guest intro — Magnus McHune, HiveMQ

01:55 What a Principal Architect actually does (customer-driven R&D)

02:40 Origins & verticals: automotive → manufacturing, data centers

03:55 MQTT + Unified Namespace as the backbone

04:50 “Level up” at ICC: Ignition 8.3, venue, Prove-It flavor

05:35 HiveMQ x Ignition: modules in development & why

06:20 Moving up-stack: data context, semantics, and insights

07:20 Building for scale: edge, site, central; reliable data movement

08:25 Common scaling pitfalls (topic design, payloads, QoS, support)

10:05 From OSS POCs to supported platforms (avoiding tech debt)

12:00 The flywheel: once UNS is live, use cases multiply

13:10 Getting started fast: containers, K8s operator, clustering

15:00 Bridging OT/IT & business needs with MQTT + extensions

16:40 Services & support: solutions, professional services, 24/7

18:05 The big three: cost, complexity, cybersecurity

19:30 Data-center patterns & data federation

21:05 Security story: TLS, auth, one-way outbound connections

23:10 Where AI/ML fits: pipelines today, context with HiveMQ Pulse

24:45 MCP & tool-calling—what to watch next

26:30 Wrap-up & where to learn more