Cloud technologies and microservice architectures are closely linked to the implementation of IIoT platforms in the predictive maintenance. Cloud computing in particular is of enormous importance in predictive and remote maintenance. Our IIoT platforms are based on established Docker and Kubernetes technologies. The aim of this scalable cloud computing approach is to provide IT infrastructures, platforms and software via the internet without the need for local installations. Both provision and use are realized solely through protocols and interfaces – usually via web browsers.
Cloud communication takes place in direct connection with intelligent IoT Gateways. This is also possible offline, which is why the capacity of the cloud can be lower and thus a cost-effective implementation can be realized. A so-called fog computing architecture can be found between the IoT Gateway and the cloud. Fog computing allows certain functions to be outsourced from the cloud near the network. The aim is to reduce the volume of data and relieve the bandwidth used by companies. This approach is relevant for the industrial Internet of Things, as many networked machines transmit data simultaneously and process it, especially in predictive maintenance. In this way, a partial evaluation of the data is carried out and prepared for transfer to the cloud. This also makes sense in terms of data security, as sensitive data does not have to be transferred to the cloud, but remains on-site.
Another aspect of the IIoT platform is the underlying microservice architecture. Instead of providing a large service, the architecture is split into separate small services, which allows a fast implementation of functionalities. In short, the IT architecture is modularized. The stand-alone services are usually designed for a single task, but specialize in the same. Taken together, the individual services result in an optimally coordinated package of services, such as module weInspect, for example. weInspect has several independent functions (which can be individually developed, tested, updated, removed and added), which together form a software for collaboration between on-site technicians and remote experts. Thus, each element of the module is a single microservice. This flexibility allows us to respond even more closely to customer requirements and implement Smart Remote Service on an individual basis.
Cloud computing and microservices are closely linked to our products such as weMonitor. The relationship between the technologies results from the product requirements. In order to identify these, various aspects have to be considered during configuration:
(1) Processing performance: In the course of predictive maintenance, the simultaneous monitoring of many machines is an elementary component. In order to meet the requirements in terms of data volume, the IIoT platform must record and correlate millions of sensor data per second. This is not a classic big data approach, but real-time data stream processing in the IoT age (Fast Data). weMonitor therefore supports the processing up to 20 different sensor signals per machine with a sampling rate up to 5 ms. These sensor signals are converted into data streams. weMonitor manages to utilize the data streams of any number of machines worldwide, display them via the webapp and visualize them.
(2) Modularity, long-term usability, flexibility and extensibility: weMonitor as an IIoT platform is designed to be operable and usable in the long term. In order for our application to be able to adapt to the changing conditions, prerequisites and customer requirements, it is necessary to rely on a modular and expandable IT environment. Therefore, the additional and future integration of further services - such as more extensive alerting, machine learning or extended analytic modules - can also be implemented. Here, too, the microservice concept is taken up, as additional modules are only coupled loosely with the basic system and flexibility is maintained.
(3) Scalability, reliability and availability: weMonitor is connected to many machines worldwide. Therefore, factors such as scalability, reliability and availability are important for a stable running of the module. Since the load situation of machines and systems varies greatly, dynamic and load-independent scaling is implemented. This must be possible to a very high degree, since hundreds of computer nodes or tens of thousands of microservice instances operate in the background. In order to ensure continued reliability and constant availability, we use mechanisms for the self-healing of the platform and the implementation of rolling updates.
Advantages of our IIoT platforms at a glance
Deployment and use without additional installations
Fog computing to reduce data load
Sensitive data remains in place
High portability
Rolling updates for continuous improvement
Cost-saving due to low resource consumption
High productivity in development
Broad community support
Fast implementation of new functionalities
Self-healing at the microservices level
Microservice-based, flexible scaling
Better maintainability through smaller, well-structured microservices
Independent development of microservices
Kubernetes is a registered trademark of The Linux Foundation in the US and/or other countries.
Back to technology overview