With the rapid development of the industrial Internet of Things, a large number of underlying data put cloud computing architecture to the test. The emergence of edge computing can take full advantage of the advantages of high-performance computing devices. Pioneer edge computing devices can be located near the side of the data source device. Through the distributed open platform integrating network, computing, storage and application core capabilities, data can be collected, analyzed and processed locally, so as to significantly reduce the processing delay caused by cloud computing, reduce network bandwidth requirements, effectively reduce the opportunities for data exposure to the public network, and protect data privacy, It is widely used in "smart communities, smart factories, public security, smart homes" and other industries.
If the manufacturing industry wants to upgrade to an intelligent factory, cameras, robots, conveyor belts and other Internet of Things devices are essential. When a large number of intelligent terminals and devices are accessed through the industrial network, the daily business data that enterprises need to calculate and process will become very large. At the same time, industrial manufacturing equipment also needs timely and rapid feedback. Cloud computing architecture is difficult to achieve real-time response, so edge computing must be used at the data generation end for preprocessing and judgment, and feedback instructions must be made quickly to ensure production efficiency.
An autonomous vehicle needs to respond to the surrounding environment at a high speed, collect and analyze a large amount of road data at any time, and then make a judgment on vehicle movement, which can be successfully processed in real time only with the help of edge computing.
More and more families use electrical equipment with data collection and networking functions. The data collected by these equipment is very private. It involves not only personal daily behavior, but also images, videos and other information. In addition to considering the bandwidth of the home network, data privacy is also important. We need to calculate the gateway as an edge node to let it consume the data generated in the home itself, Then these abstract data are combined and organically unified.