Edge computing

In contrast to cloud computing, edge computing refers to decentralized data processing at the edge of the network , the so-called edge.

description

When EdgeComputing computer applications, data and services from the central node (are data centers ) displaced away to the outer edges of a network. In other words, it is a matter of processing data streams at least partially on site (e.g. directly on the device or within a factory) in a resource-saving way, but still benefiting from the advantages of the cloud. This approach requires the use of resources that are not permanently connected to a network such as controllers , notebooks , smartphones , tablet computers and sensors . Edge computing includes numerous technologies such as sensor networks, mobile data acquisition, mobile signature analysis , peer-to-peer and ad-hoc networking. Edge computing can be used as an architectural concept for the Internet of Things (IoT), which creates a link between clearly identifiable physical objects (things) and a virtual representation in an Internet-like structure. Instead of edge computing, the terms fog computing , local cloud or cloudlet are occasionally used. With Fog Computing, however, the focus is less on the end devices and more on bringing the cloud resources closer to the applications (decentralization).

“The intelligence of data acquisition devices and sensors has grown considerably and has become more decentralized. The processing elements moved closer to the sensor. But not only the measuring devices have become more intelligent. Intelligent sensors also emerged that combine the sensor element, signal conditioning, embedded processing and digital interface (the bus system ) in an extremely small form factor or system. ”Rahman Jamal, Global Technology and Marketing Director of National Instruments (NI)

Advantages and disadvantages

The services of the Edge App significantly reduce the data volume to be transmitted and thus the data exchange and the transmission path, which reduces transmission costs and waiting times and improves the overall quality of service. With edge computing, central data centers are rarely or not at all necessary, which avoids a major bottleneck for data transfer and a potential source of error. Security is also improved as encrypted files are processed closer to the network core. When the data reaches the company, viruses , corrupted data and hacker attacks can be intercepted at an early stage. Ultimately, the ability to virtualize expands the scalability , which means that the number of edge devices in the network can easily be increased. Edge computing supports real-time requirements in the Internet of Things better than is the case in the cloud. Disadvantages of edge computing, on the other hand, arise when a very large amount of data has to be processed or stored or the computing or storage requirements are very irregular.

outlook

Edge computing plays a major role in networking the means of production in Industry 4.0 with information and communication technology. The expansion of the conventional programmable logic controller (PLC) to the edge controller should also be seen in this context.

Edge controller

The Edge Controller is a form of programmable logic controller (PLC) in which the processing elements move closer to the sensor and their data can thus be quickly analyzed and filtered. As a local controller, he is able to reduce the volume of data in decentralized data centers. The edge controller differs from the conventional PLC mainly in the greater processing ability of data, which is ensured by the use of multicore processors with higher computing power.

"Edge controllers are devices that enable the use of cloud technologies in production halls and machines." Alexander Bergner, Product Manager Industrial IoT at TTTech Computertechnik

See also

• Embedded system
• Mobile Edge Computing (Engl.)

Web links

• National Instruments White Paper: http://www.ni.com/white-paper/52926/de/
• ÖFIT trend show: Edge Computing , Competence Center Public IT

Individual evidence

1. ↑ Wolfgang Wanner: Cloud versus Edge Computing. Funkschau, June 6, 2016, accessed on June 21, 2017 . 
2. ↑ Edge computing vs. fog computing. General Electric, accessed June 4, 2018 . 
3. ↑ Andreas Knoll: Edge controller instead of PLC? Markt & Technik, November 3, 2016, accessed June 21, 2017 . 
4. ↑ Frank Sempert: EdgeComputing for the Internet of things. (No longer available online.) Computerwoche, April 15, 2016, archived from the original on April 18, 2017 ; accessed on June 21, 2017 . Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice.  
5. ↑ Andreas Knoll: Does the edge controller replace the PLC? computer-automation.de, November 23, 2016, accessed June 21, 2017 .