What Is Edge Computing?

Edge computing originated in the media field. It refers to an open platform that uses the core capabilities of network, computing, storage, and applications on the side close to the source of data or data, and provides the nearest-end services nearby. Its applications are launched on the edge side, resulting in faster network service response, meeting the industry's basic needs in real-time business, application intelligence, security and privacy protection. Edge computing is between, or on top of, physical entities and industrial connections. And cloud computing can still access historical data for edge computing. ^[1]

Edge computing is not a new word. AKAMAI, a provider of CDN and cloud services for content delivery networks, has been working with IBM on "edge computing" since 2003. As one of the world's largest distributed computing service providers, it was responsible for 15-30% of global network traffic. In one of his internal research projects, he proposed the purpose of "edge computing" and solved the problem, and provided edge-edge-based services on his WebSphere through AKAMAI and IBM. ^[1]

Abroad

The rapid development of global smartphones has promoted the development of mobile terminals and "edge computing". The intelligent society in which everything is connected and everything perceives is accompanied by the development of the Internet of Things, and therefore edge computing systems have emerged.

In fact,

In China, the Edge Computing Alliance ECC is working to promote the integration of three technologies, that is, the integration of OICT (Operational Operation, Information Information, Communication Technology). The calculation object mainly defines four fields. The first is the problem of the device field. ^[1]

Automation is in fact a "control" at its core. Control is based on "signals" and "calculation" is based on data. More meanings are "strategy" and "planning". Therefore, it focuses more on "scheduling, optimization, and path". Just like the national high-speed rail dispatching system, each additional train reduction will trigger the adjustment of the dispatching system, which is a time- and node-based operation planning and planning problem. The application of edge computing in industry is more of this kind of "computing."

Simply put, traditional automatic control is based on signal control, and edge computing can be understood as "information-based control".

It is worth noting that although edge computing and fog computing are talking about low latency, their 50mS and 100mS cycles are still for "control tasks" such as 100S of high-precision machine tools, robots, and high-speed graphic printing systems. There is a very large delay. The so-called "real-time" of edge computing, from the perspective of the automation industry-unfortunately, is still classified as "non-real-time" applications.

Edge computing is a technology developed in the context of high bandwidth, time sensitivity, and IoT integration. The concept of "Edge" was indeed mentioned earlier by automation / robot manufacturers such as ABB, B & R, Schneider, and KUKA. The original intention is to cover those "IT resources close to users and data sources." This is a design that extends from traditional automation vendors to IT vendors. On April 5, 2016, Schneider claimed that it could define the physical infrastructure for edge computingalthough the main idea was its "microdata center" concept. While other automation vendors mention computing, they all show a trend of integration with IT, and have the concept of edge and ubiquity.

In fact, IT and OT are infiltrating each other. Automation manufacturers have begun to extend the IT capabilities of their products, including big companies such as Bosch, SIEMENS, and GE. In terms of information and digital software platforms, they also include companies like Bega Lai, Rockwell, etc. all provide products and technologies for basic IoT integration and integration of Web technologies. In fact, IT technology has also begun to integrate bus interfaces, HMI-enabled products in its products, and industrial field transmission equipment gateways, switches and other products.

IoT is regarded as a rapidly growing field in the future, including the most cutting-edge Internet-based technologies that have appeared. Qualcomm has proposed the Internet of Everythingwhich can be called IoX. Therefore, a new industry pattern is emerging. In terms of the boundary definition of the ECC, Huawei's main goal is to provide computing platforms, including basic network, cloud, edge server, transmission equipment and interface standards, while Intel and ARM provide edge. The computing chip and processing power are guaranteed. The ICT Institute plays the integration of the transmission protocol and system implementation, while the Shenyang Automation Institute and iSoftStone play the role of practical applications.

However, edge computing / fog computing must be implemented. Especially in industry, "application" is the core issue. The so-called integration of IT and OT emphasizes the application on the OT side, which is the goal of the operating system. .

In the industrial field, edge application scenarios include energy analysis, logistics planning, and process optimization analysis. As far as production task allocation is concerned, the optimal equipment scheduling for production needs to be performed according to the production order. This is the basic task unit of APS or generalized MES and requires a lot of calculations. Whether these calculations depend on the software platform of a specific MES manufacturer or an "edge computing" platform-an analysis platform built on the basis of Web technology, will not have much difference in the future. In a sense, the MES system itself is a traditional architecture, and its core can be either a dedicated software system, or it can exist on the cloud, fog, or edge side. ^[2]

In such application scenarios, in general, the respective divisions of labor in the entire application of intelligent manufacturing and industrial Internet of Things are as follows.

Automation manufacturers provide "collection", including the role of data sources. This is the use of native "information" such as machine production, status, and quality generated by distributed I / O acquisition, bus interconnection, and control machines.

Connection architecture for edge computing (orange)

ICT vendors provide "transmission" for industrial connectivity. Because in terms of how to provide data transmission, storage, and computing, ICT vendors have their traditional advantages, including cost, and have the advantages of cloud platforms.

The business experience and knowledge of traditional industrial enterprises provide the basis for "analysis" for analysis software (independent or internal) vendors. Understanding of these business processes is still essential. The ultimate goal of collaboration in the industrial chain is still to resolve the core issues of "quality, cost, and delivery."