Artificial intelligence has always been popular in recent years the field of science and technology circle, the country is in the past few years, such as the birth of the desert, as the technology, science and technology Shang, very Chain Technology Video ++, in accordance with plans of outstanding science and technology start-ups. But with the arrival 5G, artificial intelligence and Times of things, traditional cloud computing technology has been unable to meet the AI terminal “big connections, low latency, high bandwidth ‘needs. Although the ability of cloud computing more powerful, but the face of a large number of data related to personal privacy, the traditional cloud computing is still not effectively support application service program based on the Internet of Things, and the edge of the cloud at the edge of the big-data processing computing era just can solve these problems.
edge cloud computing is what the edge is the ability of the core and edge of cloud computing technology-based computing, cloud computing platform to build on the edge of the infrastructure. Forming an edge position computing, networking, storage, security and other capabilities fully elastic cloud platform, and to form a “side edge of the cloud trisomy synergistic” end to end architecture and technology center and cloud things terminal, by forwarding the network, storage, calculating, intelligent data analysis processing work on the edge , reduce the response delay, reduce the cloud pressure, reduce bandwidth costs, and the entire network for scheduling, power distribution, etc. operator services cloud. In simple terms, the edge computing refers to the network edge or near the source of the data, combines networking, computing, storage, and application processing capabilities of distributed platforms, the nearest to provide intelligent services. Edge computation will be appreciated that the reverse operation as a cloud, the cloud is emphasized that the computing and storage capacity and the like, or a desktop terminal edge of the end concentration, and this edge computation computing and storage capacity sucked back to the edge of the sink. Edge computing causes major cause of edge computing or lack of cloud computing services, cloud computing method they use centralized management, which allows cloud services to create a higher economic efficiency, but in the context of the interconnection of all things, application services require low latency, high reliability, and data security, and traditional cloud computing can not meet these needs. Was the first networked environment , the edge device generates a large amount of data in real time, are reaching the cloud performance bottleneck, according to IDC, to 2020, the total amount of data will be greater than 40ZB, with increasing the data amount of edge devices, the network bandwidth is becoming a bottleneck in another cloud. Second, when a userWhen using e-shopping sites, search engines, social networking, user’s private data will be uploaded to the cloud center, which contains a user privacy data, if directly to the video data is uploaded to the cloud data center, transfer video data will not only bandwidth resources, but also increases the risk of disclosure of user privacy data points, the edge computing model provides precisely the amount of good privacy protection mechanisms for such sensitive data. Finally, the cloud data center energy consumption, as more and more user applications running in the cloud computing center, the future demand for large-scale data center energy consumption will be difficult to meet, in order to solve the energy problem, the edge computational model proposed some computing tasks running on the original cloud data center to decompose, then migrate the decomposition of computing tasks to the edge node for processing, in order to reduce the computational load cloud computing Center control Engineering Copyright [123 ], in order to reduce energy consumption purposes. Edge computing scenarios edge computing scenarios from the upper cover and covering the entire network can be divided into two types of local coverage. The core cover the entire network-based applications from the coverage of the edge node is on two levels and carrier network to ensure the nearest calculated (e.g., the CDN, live video, the edges of the dialing test / monitoring services), or based on a sufficient number the network node link optimization. Core type applications require local coverage localized edge node, i.e. the access edge node for a distance close enough (live video media stream push to the nearest edge node transcoding edge node directly, transcoded media flow distribution to the CDN edge node, when a user accesses the nearest return content based on the service edge node, uplink and downlink content and live streaming push transcoding process so do not come back to the central processing, greatly reduces business latency, improve interactive experience, while edge processing architecture to save bandwidth cost is also very obvious Summary: AI must be carried out over the past operational data analysis and algorithms rely on a strong cloud computing capabilities, as well as with the advent of sophisticated new applications of technology , commercial digital concept gradually deepened people’s thinking, chip capacity has been increasing, edge computing platform matures and begins AI gives greater capacity to assist the initial screening data analysis, real-time response and other plant equipment, in industry, smart city, video identification service can make further Lifting edge will be AI cloud computing, networkingThe key component of the fields, will also be greater development, covering potential customers and the scene will continue to emerge. As more and more edge cloud computing scenarios, user needs change in the future will be the focus, if done well Control Engineering Copyright , the next edge cloud computing cloud computing than traditional low more costs to realize the project.