Secure Edge Computing for IoT-Systems

For many applications it becomes clear that data collection in the cloud is not the optimal solution. There are currently two competing approaches to data storage and processing: Cloud and Edge Computing. Digital applications, in particular, generate large amounts of data that are usually not actually needed to this extent. In addition, some processes require real-time information so that they can respond quickly to various events. When applications need data on an economical scale immediately and above all securely, it is questionable whether data storage in the cloud is the most suitable solution.

Edge Computing only transfers data that is actually needed in the cloud to optimize processes. This architecture makes it easier to meet security requirements and the desire for data economy. The availability of systems, low latency times as well as data backup and encryption are easier to implement.

A gateway collects all incoming data and stores it internally. It then selects the data and sends the relevant information according to individual specifications for storage on local servers or, if necessary, in the cloud. Services with analysis functionality prepare this data for further processing.

• How do I manage the multitude of different interfaces?
• How do I link the different island solutions with each other?
• How do I integrate new application programs?
• How do I take IT security into account in my system?

Challenge when using an AI (e.g. for predictive maintenance)

Accurate data is needed so that subsequent processes can be executed correctly. Incorrect or inaccurate data leads to wrong decisions. Companies therefore ask themselves the question when using cognitive intelligence: How do I get the right data?

What data is actually required?

In many applications, the machine control data is often not sufficient to assess the quality of a produced product. With additional information, scrap can be reduced and the life of equipment can be extended.

A suitable information architecture is often required in real time

An optimized data management which brings together different sources, stores the data centrally and can process structured and unstructured data, is required in the IoT. Metadata management, which provides proof of data origin, storage location and system status, is also necessary.