EdgAIR: The cloud for machines

Deutsche Telekom T-Systems now offers a platform with extremely low latency for secure and powerful edge computing. This will enable companies to use innovative real-time applications in production and logistics, including automated guided vehicles. Not to mention solutions for augmented and full virtual reality. Called EdgAIR, the platform works directly on site. This means that the data is not sent via a central computer center.

EdgAIR is designed to be like a private cloud of things, making it highly secure. Customers can link it up to a campus network. EdgAIR can then wirelessly transfer data between devices and the platform in near real time. Experts consider "near edge" to be less than ten milliseconds. The so-called "far edge" allows latency times between ten and 40 milliseconds. "In combination with Time Sensitive Networks (TSN) and a very good Service Level Agreement, our managed platform EdgAIR can meet requirements such as real-time control of machines in a production hall," explained Albert Kroisleitner, Senior Product Manager Edge Platform & Solutions at T-Systems. Each EdgAIR platform is also delivered with Edge Analytics. This tool enables data scientists and software developers to evaluate information on site. And thus quickly adapt processes and models (code to data).

Deutsche Telekom has already built its first campus network for customer OSRAM. Such networks form the basis for EdgAIR. Currently running on LTE, they will soon switch to 5G. T-Systems offers the platform tailored to various sectors. These include production, logistics, building automation, automotive and energy. Furthermore, for enhanced security, the solution runs on a closed company network and is completely self-sufficient. However, it can also be combined with a public cloud accessible via the internet and thus used as a hybrid cloud. 

Technical details

EdgAIR is based on OpenStack. The open source software avoids vendor lock-in. Customers can connect their IoT applications via ready-made connectors. The applications themselves run as virtual machines or in docker containers as microservices and are thus independent of the underlying infrastructure. This allows administrators to control the microservices individually in the event of problems or anomalies. The entire system does not have to be shut down. The containers are coordinated via Kubernetes. T- Systems offers the platform in various sizes and designs (pole mount, ruggedized rack, standard rack).