By Bill Lydon and Azure Security News
Microsoft has been a major software technology supplier to the industrial automation industry since the mid-1980s and is now providing a comprehensive manufacturing business architecture to reap the benefits of digitization with Azure Industrial IoT. Comprehensive industrial digitalization platforms empower manufacturers to integrate entire manufacturing business improving & production performance and profits based on their knowledge and know-how without having to design and maintain the entire platform.
Microsoft has been engaged in industrial automation for many years, starting with DOS applications and dramatically accelerated with the introduction of the Windows operating system in 1985. Wonderware’s 1987 introduction of InTouch Windows-based HMI added significant features and open interfaces to IT and business systems. Microsoft Windows provided a rich environment that spawned a large ecosystem of developers providing software for a wide range of industrial applications including databases, analysis, advanced control, Manufacturing Execution Systems (MES), batch management, production tracking and historians. Windows also provided the platform for development of OPC which significantly simplified software drivers for industrial networks and equipment interfaces.
During the 25th annual ARC Industry Forum Microsoft’s Erich Barnstedt, Chief Architect, Standards & Consortia, Azure IoT presented with a focus on, “Don’t forget what you already know!” and leveraging OPC UA.
Step by step to digital transformation
Barnstedt described a step-by-step digitalization approach.
- Make sure you have connectivity options for your assets including PLC and controllers that can be connected to a local on premises gateway.
- Install gateways and connect to all assets.
- Use online dashboards to gain visibility and understand what is happening.
- Capture time series/historical data and model relationships
- Apply machine learning and predictive algorithms.
- Define automatic responses creating a digital feedback loop for optimization and continuous improvement.
Barnstedt stated, “When we first started working in the industrial space over 15 years ago, it came back to us over and over again: manufacturers were complaining they were locked into certain vendors that were dictating prices and keeping data in silos. When we set up the AZURE IoT team, we wanted to build the product in a way that avoids vendor lock-in.” Barnstedt emphasized these points about vendor lock-in:
- You are forced to use a closed source SDK in your product.
- You are forced to use a proprietary interface to communicate.
- You are forced to use a proprietary communication protocol.
- You are forced to use a vendor specific device data description/model.
- You can no longer use your private/on premise data center.
- You are forced to use hardware/software from a single vendor.
Barnstedt believes this “makes you forget what you already know! This locks you into a vendor, creates extra configuration and implementation costs and limits flexibility. Based on this, the Azure IoT team created a set of business and technology principles.”
|Business Principles||Technology Principles|
|Modular Industrial IoT PaaS platform|
(Managed and hyper-scale)
No proprietary solutions
Support for micro – services and manage container services on the edge in in the cloud
|Secure by default|
(Follows Microsoft’s Security Development Lifecycle- SDL)
|Protocol independent, platform-independent|
Works just as well with AMQP, MQTT, and HTTPS as well as Linux and Windows operating systems
|Scales through Ecosystem|
(World’s largest partner network to solve any complex problem)
|Uses an open data/information model|
Along with open-source tools
|Largest hardware OEMs ecosystem for gateways||Based on Open Industrial Interoperability Standards|
Compatible with Reference Architecture Model Industrie 4.0 (RAMI) [https://www.isa.org/intech-home/2019/march-april/features/rami-4-0-reference-architectural-model-for-industr ] and using OPC UA
|Most certified/compliant cloud||Non-intrusive|
connect your machines without modification
|Partnership with OT companies|
(creating a win – win for IT/OT merger)
|Most data center regions & world’s largest network|
(Plus, sovereign clouds in US, Germany & China)
Barnstedt elaborated that most importantly vendor lock-in does makes you forget what you already know because the data and data models you’ve already created cannot be used with multiple vendor systems and you have to start all over configuring data models describing your existing data in a new format and a new model which cost you time and money and locks you into that particular vendor.
Azure IIoT is based on open source, open platform, open standards and open data models. The Azure IIoT architecture fully supports OPC UA and data models from edge to enterprise and cloud.
OPC UA & MQTT
Barnstedt clarified a common misconception that OPC UA in MQTT are competitive technologies, noting they are complementary.
- Fact: OPC UA leverages MQTT (see part 14 of OPC UA specification)
- fact: OPC UA leverages JSON (see part six of OPC UA specification)
- fact: OPC UA leverages both technologies since 2017
- Fact: OPC UA is an IEC standard supported by hundreds of manufacturers, cloud providers and automation companies
- Fact: OPC UA supports and extensible model for defining new information models via companion specifications for industry use cases.
Lightweight cloud Telemetry is common with OPC UA as a JSON payload on top of MQTT including data types.
AZURE open industrial IoT stack
Microsoft provides the open manufacturing platform and infrastructure based on open source software available on GitHub The Azure open Manufacturing platform based on OPC UA which are open source on GitHub and include OPC Publisher, OPC twin, OPC Discovery Service, IoT Edge and IoT Hub.