Since TTTech was founded in 1998, Georg Stöger has supported the company as a product developer, project engineer, system architect and trainer in the field of digital communication and control systems. As part of this role, he supports a wide range of customers and has thus gained a broad overview of the requirements of a wide range of industries and sectors. Since 2019, he has been working at the subsidiary TTTech Industrial as a technology consultant for product development and customer projects, but also as a 'technology evangelist' for publications and events.

The Industrial Internet of Things is generating more and more data that needs to be managed. This data complexity needs to be mastered. What role does edge computing play in this?

Stöger: The management of complex OT data is one of the most important tasks of edge computing and encompasses many aspects:

Data acquisition is possible via various interfaces - on the one hand via various fieldbuses, and on the other via interfaces defined by open (e.g. OPC UA) and de facto (e.g. S7) standards. In the case of large volumes of data and/or fast data cycles, this step alone can pose a challenge for the edge infrastructure.

Aggregation, filtering and classification with statically defined algorithms or artificial intelligence methods, which are also increasingly suitable for real-time data stream analysis.

Detection of anomalies in the data in order to write logs, trigger diagnostics or trigger an alert via email/SMS, among other things.

Visualization of the data in different dashboards directly at the Edge or in remote access.

Storage of relevant data at the edge and in the cloud. The architecture of the edge solution has a significant influence on which data is relevant in each case and what storage space and bandwidth restrictions are faced.

Security also plays an important role when it comes to data management, as data increasingly needs to be provided with metadata for access authorizations and restrictions at the edge before it is forwarded and processed in different applications.

At the edge, sensor data is preselected, software applications are processed or even AI calculations are carried out. What should users bear in mind with regard to the edge software landscape? Which technologies should/must be taken into account?

Stöger: Flexibility and user-friendliness compete with each other here: on the one hand, you want to be able to use as many different types of applications as possible at the edge - innovative applications and autonomous systems based on machine learning, but also user interfaces, diagnostic applications, cloud connectors, security applications, etc. should be able to run at the edge. On the other hand, human users are usually not IT specialists and are not familiar with complex software installation and maintenance tasks.

TTTech Industrial wants to exploit the potential of automation technology.

© Pixabay / CC0

It is therefore important to present as flexible a set of application options as possible in a package that is as easy to use as possible and to make it usable for both local and remote management. In our view, this requirement can be comprehensively addressed with a combination of virtualization (primarily for legacy applications and HMIs), Docker (for all types of data processing) as well as programmable controllers in accordance with IEC 61131 and process controllers in accordance with IEC 61499.

It is up to the user and their needs to decide whether these very different software applications should run on a single universal hardware platform or on different hardware platforms depending on the task. It is therefore essential that the edge platform used can support both options - this also gives the user the greatest possible flexibility in the selection of applications and in setting up and expanding the appropriate edge architecture.

The hardware requirements grow with the tasks. What should users look out for when selecting the right edge computing hardware?

Stöger: First of all, many users already have appropriate hardware (industrial PCs and industrial servers) in use that they would like to continue using when introducing or expanding their edge architecture; the question here is rather which edge software solution best supports this hardware infrastructure already in use.

The criteria for selecting new industrial edge hardware are similar to those for other industrial hardware: robustness for long-term operation under harsh environmental conditions (dust, temperature, vibrations, etc.) and a good cost-benefit ratio are at the top of the list. It is generally difficult to estimate how many CPU cores, network connections, RAM and storage space are required, but for more complex edge applications with virtualization (hypervisor) and containers (Docker), newer CPU generations with four to 16 or more cores and correspondingly many gigabytes of RAM are definitely relevant.

As edge management platforms support the central administration of applications through to application orchestration, it is most cost-efficient in the medium term to use only a few (but more than one for availability reasons) very powerful computer platforms on which the entire edge application landscape runs. However, this is not yet the norm - we currently mainly see IPCs in the two- or four-core performance class at the Edge.