The concept of Edge Computing has suddenly been on fire in the past year or two. Although in the industrial control industry, everyone actually does not understand or accept the concept of cloud computing. In addition, it is somewhat interesting that edge computing is not like cloud computing or fog computing. It is not an application-driven one. It is not presented by the "Jobs" personage or leader company of the IT industry. concept.
For the industrial control community, basically the concept of edge computing belongs to an exotic. Carefully pondering the meaning of the edge or its English "Edge", what is the edge, is actually relative to the overall structure of IT, the general idea in the IT field is from the top down, since the edge computing is located near the industrial scene, The location close to the data source, which is very natural, is at the end of the overall IT architecture, at the edge. The concept of edge computing comes from this. Based on this idea, it is easy to understand that edge computing complements cloud computing. It complements cloud computing's deficiencies in industrial applications and meets the needs of industrial users. The basic purpose is also to promote the convergence of IT and OT. To respond to the big development strategy of industrial Internet of things.
Benefited from Huawei's drive, domestic operations in the edge computing area are relatively fast. As early as January 2016, Huawei led the establishment of the Edge Computing Industry Alliance with five other companies including Intel and ARM. In all fairness, look at the list of the most core members of the Alliance including the six members of the board of directors, including Huawei. The industry control community, especially the users, may feel that the six members of the board of directors are not suppliers in the traditional industry. The calculation of the edge of the "Chang Teng" is still far away from us.
However, the progress of edge computing in the industrial field seems to exceed expectations. In November last year, led by Mitsubishi Electric, is also the establishment of the Edgecross Alliance 6 start-up members, Omron is also among the first members. Taking into account Mitsubishi Electric, Omron's position in the field of industrial control, it also means that the industrial control industry for the recognition and introduction of the concept of edge computing. In January this year, Fanuc, who has always been low-key, has also formed a joint venture with Hitachi and FSN. The three parties have complementary advantages and focus on the development of intelligent edge systems. It seems that marginal computing is driven by the wave of technologies such as industrial internet of things and artificial intelligence. It is not so far away from us.
In conjunction with cloud computing, edge computing can also be responsible for a number of computing and analysis functions. At present, the consensus in the industry is predictive maintenance analysis. Compared with cloud-based analysis, edge-based predictive analysis can achieve almost real-time performance, which is very attractive for industrial control. What needs to be clarified is that at present, the definition of edge calculations does not involve the processing of industrial control logic or algorithms, which still belong to the category of industrial controllers.
In addition, one thing may be easily overlooked by everyone. Edge computing is part of the IT system. It can also enjoy the benefits of the rapid development of IT technology, while avoiding the inherent shackles of traditional industrial control. For example, 5G wireless communication technology that is close to commercial use, low-power low-cost LPWAN technology, very innovative SDN network architecture, TSN time-sensitive network technology, and so on. These new technologies can improve the vitality of edge computing.
The above description of edge computing still belongs to the category of IT. As for the concept of edge computing will not further introduce the traditional control system into the OT category, the mainstream thinking that the introduction of the concept of edge computing, the control system in the system architecture, software and other areas of innovation will get better performance; current international There are also many companies and organizations in the industry who are exploring and studying such innovative solutions. Of course, it takes a long time to develop and verify.
Extended reading: Now that the artificial intelligence AI technology is as popular as any new IT technology, we can't deny that the artificial intelligence depends on the powerful support of the cloud computing platform. Many AI's specific needs depend on the cloud computing platform. Edge computing is done, however, AI is still limited in the application deployment process to the edge of the cost of computing and equipment can only analyze the ability and many other aspects.
Whether it is from the current national policy support or the promotion of business applications, we can all see how artificial intelligence is now in the end. According to the research data of authoritative market analysis agencies, the global artificial intelligence market will grow in the future. The average annual growth rate reached 15%. By 2030, artificial intelligence will boost global GDP growth by about 12% to nearly $10 trillion. Such a huge market size is enough for more and more companies to join them.
What exactly is edge computing? Because edge computing is important for artificial intelligence and cloud services, we need to understand exactly what edge computing is. The so-called edge calculation is an open platform that integrates network, computation, storage, and application core capabilities on the edge of the network close to the physical device or data source, and provides edge intelligent services nearby to meet fast connectivity, real-time services, data optimization, and application intelligence. Key requirements technologies such as security and privacy protection.
Edge computing is more focused on the analysis of real-time, short-cycle data. Edge computing can very well support real-time intelligent processing of data and the specific execution of tasks by local businesses. Due to its closeness to the data source, edge computing can perform operations on the local network. The data touched does not need to be uploaded to the cloud, reducing the time and bandwidth of the data to and from the cloud and the local cost.
Edge computing is now widely used in the field of automated driving. We all know that due to safety considerations, data transmission and interaction in the field of automated driving must minimize data delays as much as possible, because if not done, the resulting The problems and results may be very serious. In this area, edge computing will undoubtedly have advantages over cloud computing. Therefore, it is not surprising that edge computing has become the focus of AI-related chip manufacturers and device integrators, and it has directly driven the rapid rise of edge computing in 2017.
The other side of edge computing <br> Although edge computing has a bit of what we said before, due to limitations in smart analytics capabilities and legal compliance of data applications, edge computing is still in a slow development. In the stage, the unit price of the AI ​​equipment using edge calculation is relatively high. After the general terminal electronic products are equipped with AI chips and the storage space is improved, the cost of the equipment will be greatly increased, and the overall cost performance of the terminal is not high. Many companies will obviously hinder the popularization and promotion of related products after they are faced with such a high cost of technology.
The computing power of edge computing terminals can play a role of "smart" analysis. According to the forecast of the International Data Corporation, more than 50 billion terminals and devices will be networked by 2020, and more than 50% of the data in the future will need to be analyzed, processed, and stored at the edge of the network.
In areas such as medical care, autopilot, and national defense that rely more on edge computing technology, due to the difficulty of data acquisition and the low degree of loosening of laws and regulations, the risk will actually become higher when the artificial intelligence is actually imported, and the tolerance for risk will decrease. .
What cloud computing can do for
us <br> For artificial intelligence AI, there is still a need for an intelligent calculation method for the processing capacity and rendering ability of massive data, and cloud computing can be integrated through the use of different data centers. The algorithm is integrated to help the majority of users achieve data processing level service requirements.
At the same time, for the edge computing that is still in its infancy, cloud computing technology can effectively reduce the developer's threshold and enable users to quickly convert artificial intelligence services into productivity in a low-cost way.
The rise of edge computing has not only led to sales of AI chips, but has also led to further expansion of the relevant hardware market. It can be foreseen that once all terminal equipment can implement edge computing, its market will surely go far beyond cloud computing.
Special Cable
Special Cable,Cable Special,Special Electrical Cables,Special Cables Industries
Shenzhen Bendakang Cables Holding Co., Ltd , https://www.bdkcables.com