Since we started the Azure Industrial IoT team six years ago, one topic that tends to dominate the business decision-maker meetings I’ve participated in: vendor lock-in. Vendor lock-in is the inability to pick the best solution and instead be forced to pick the solution from a vendor you have used in the past to maintain compatibility with existing systems.
If you’ve worked in the manufacturing space for a long time, this will not come as a surprise. For decades, manufacturers had to build digital manufacturing solutions in closed ecosystems—most of which only offered expensive, proprietary interfaces and kept your data in silos such as proprietary database systems.
Many manufacturers want to break free of this lock-in and see their digital transformation journeys as an opportunity to unlock the innovation of the full manufacturing community and partner ecosystem. Microsoft Azure IoT aims to simplify IoT for mainstream adoption and is helping to drive openness and interoperability in industrial IoT and manufacturing data systems for the future. When making key decisions on technology, we recommend manufacturing company decision-makers look for these telltale signs so as to avoid being locked in—and ultimately simplify their journey to successfully adopting IoT solutions.
5 questions to guide Industry 4.0 vendor selection
I’ve developed five questions that can be asked about a proposed industrial Internet of Things (IIoT) solution to identify if it will lock you in. If you can answer yes to any of the questions below, you may want to reconsider using the proposed solution:
- Are there any proprietary interfaces in the solution?
If the interface you need to use to access data or systems is not based on a standard endorsed by an independent organization like IEC, ISO, DIN, or Oasis, the solution will lock you in.
- Are any proprietary communication protocols being used in the solution?
Additionally, if the solution uses a proprietary communication protocol that is not openly published and available without royalties, the solution will lock you in.
- Does the solution use proprietary data formats?
If the solution encodes data in transit or at rest via a proprietary data format, you will need to transform the data to take it back out—if you are even allowed to do that. Transforming data can get expensive for large data sets, too.
- Does the solution store data using a proprietary data model?
If you need to model or remodel your data using a proprietary data model, it will be hard to map the data to another model you may be using.
- Does the solution dictate where it is run?
If the solution can only be run in a single environment, in a specific data center, or on specific hardware only available from the solution vendor, for example, it will lock you in.
Leading the pack on open standards
Azure Industrial IoT has been named the leading IIoT platform by Gartner, Forester, and IDC. This is due, in part, to our relentless dedication to avoiding vendor lock-in by embracing the innovation that takes place across the industry. We accomplish this through openness based on our open cloud platform, our open-source implementation, our support for open machine data models, and most importantly, our dedication to leveraging open industrial interoperability standards like the Open Platform Communications Unified Architecture (OPC UA).
Many technologies offered for IIoT solutions claim to be open or using industry standards or both. On closer inspection, however, it’s clear they are not all that open or are not based on a standard (i.e., endorsed and maintained by a standards organization). To make things worse, it is often hard to identify if a certain organization is really a standards organization, as many people confuse the word “standard” with “open.” Just because a certain technology or solution claims to be open doesn’t make it a standard. An easy way to recognize a standard is to see if it has a publication number associated with it—for example, 62541 for the publication number assigned to OPC UA by the IEC standards organization.
The Azure Industrial IoT platform uses OPC UA and was the first cloud platform to do so. It also uses the OPC UA machine data model, OPC UA’s open communication interfaces (both client/server and PubSub), as well as its data formats, together with OPC UA’s standard communication protocols mappings TCP/IP and MQTT.
Advantages to OPC UA standard
One way to know you have made the right technology choices is when everyone else starts comparing their solution to yours. Unfortunately, comparisons are not always easy and often include technology that will again lock you in, sometimes on purpose and sometimes by accident. This can be confusing for those not working daily in the Industry 4.0 space.
The number of adopters for a certain technology also can be important. OPC UA is endorsed by over 800 companies, including the top seven automation companies and the top three cloud providers. The standard is maintained by a nonprofit organization and is free of royalties, even for commercial use. The OPC Foundation offers an open-source reference implementation on GitHub contributed by Microsoft that can be used by OPC Foundation members free of charge for commercial use and also free of charge by nonmembers for evaluation purposes.
While Microsoft Azure Industrial IoT makes the OPC UA machine data model available in the cloud, other vendors argue OPC UA has its place to pick up data from machines but then require transforming it into a proprietary data model for cloud analysis. This results in losing the standardized OPC UA data model and such a process can waste time and money spent on creating the existing OPC UA model.
Learn more about Industry 4.0
Again, we believe in openness and avoiding lock-in for customers in our ecosystem. This helps to create devices, systems, and technology within IIoT that work together for our customers.