Reshaping Industrial Data Architectures With UNS and MQTT

In a recent episode of the Augmented Ops podcast, we explored the rapidly evolving landscape of industrial data architectures with Vatsal Shah, Founder and CEO of Litmus Automation. Titled "MQTT, Unified Namespace, and The New Industrial Data Stack," the discussion with Shah takes a close look at the architectures like Unified Namespace (UNS) and communications protocols like Message Queuing Transport Telemetry (MQTT), which are foundational to the modern data infrastructures that manufacturers demand.

Drawing from his experience as an automation engineer working with many of the incumbent vendors in the space, to eventually founding Litmus and building a new set of tools, Shah provides insights on the data integration challenges faced by many manufacturers. In particular, he highlights the urgent need to bridge the Information Technology (IT) and Operational Technology (OT) worlds with an open, interoperable data architecture.

Shah explains the significant limitations he experienced when working with the existing tools from legacy vendors in the industrial automation space, which ultimately led him to start Litmus. The issues he encountered often centered around the difficulty of integrating systems with components spanning different platforms and vendors. Shah recounts that while working on an oil and gas project, he and his team were faced with the challenge of unifying systems from Emerson, Honeywell, Fisher, Rosemount, and Siemens to all share a single database. This required a level of integration across multiple different systems that the existing tools could not achieve efficiently.

The only solution available to them at the time was to write custom code in Visual Basic, a process that was not only time-consuming but also lacked the scalability and flexibility required for modern industrial operations. The existing solutions all fell short in making the OT data readily available to IT and other systems in a streamlined manner.

To Shah, at that time creating a company was secondary to solving the technical challenges: “I didn't know if I was going to found the company, but all we wanted to do was just go ahead and solve this problem.” He eventually went on to found Litmus Automation, with the goal to create a connectivity platform that would offer a more seamless integration experience, enabling organizations to leverage their OT data more effectively across different systems and platforms.

Shah stresses the critical role of Unified Namespace, or UNS, in modernizing industrial data architectures — highlighting its pivotal role in bridging the gap between IT and OT. “OT has a mess. IT needs data right now. There has to be a common ground between both of them,” Shah says as he explains the need for UNS.

Unlike the traditional ISA-95 or Purdue models, the UNS leverages a hub and spoke architecture with a central message broker, to which everything from shop floor machines to QMS and ERP systems can connect over an open communication protocol. This eliminates the need to create dedicated point-to-point connections anytime you need to exchange data between two nodes in the namespace, and enables a much more flexible and scalable infrastructure.

Shah believes that manufacturers need an open, interoperable framework that facilitates data exchange and integration across a wide range of systems. With the advent of UNS, manufacturers can integrate new systems quicker, while also empowering citizen developers by making it easier to access data from any part of the organization.

As he explains, with a Unified Namespace, “OT decides what type of data is going to get pushed, how it is going to get organized. IT decides how it's going to get consumed, how it is going to be secure, and how it is going to have proper context that all of the smart applications require.”

Despite the revolutionary potential of this architecture, Shah acknowledges the challenges that persist due to the limitations of the data transport protocol it leverages: MQTT. The protocol is open, lightweight, publish-subscribe, and can be configured to report by exception, making it ideal for UNS use cases. However, he argues that the MQTT spec must continue to be refined in order to support Unified Namespace implementations at scale.

He notes that “MQTT came a long way…But there's a long way to go on the top of it." In order to meet the demands of enterprise-scale infrastructure and match the capabilities of data stream processors like Kafka, Shah argues that the protocol needs enhancements in resilience and scalability. He believes that this is what causes IT organizations to perceive MQTT as not yet offering the comprehensive, future-proof foundation required. One of the further developments of MQTT that has aimed to meet the demands of manufacturers is the Sparkplug B specification.

Shah explains that While MQTT is data-agnostic at the transport layer, Sparkplug B introduced a structured format for organizing data which can make integration more straightforward. However, he also pointed out the limitations of Sparkplug B, particularly its narrow scope. While it is well-suited for organizing standardized data from edge devices and SCADA systems, “the moment you go for more analyzed data, derived data, contextualized data, metadata, then you have to think a little bit beyond Sparkplug B,” Shah argues.

Ultimately, his view of the specification is that Sparkplug B’s less flexible payload structure makes it inadequate for the wide range of use cases needed to bridge OT and IT in the way that a Unified Namespace aims to achieve. Further development of data transport protocols will be required for organizations to be able to fully leverage the benefits that the UNS architecture promises.