How edge computing is hierarchically deployed

Edge computing refers to an open platform that integrates network, computing, storage, and application core capabilities on the side close to the source or data source, providing near-end services. Its applications are launched on the edge side, resulting in faster network service response, meeting the industry's basic needs for real-time business, application intelligence, security and privacy protection. Edge computing is between physical entities and industrial connections, or at the top of physical entities. In the cloud computing, historical data of edge calculations can still be accessed.

Edge computing is not a new word. As a provider of content distribution network CDN and cloud services, AKAMAI worked with IBM on "Edge Computing" in 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 its internal research projects, the purpose and problem of "edge computing" was proposed, and AKAMAI and IBM provided Edge Edge-based services on their WebSphere.

For the Internet of Things, breakthroughs in edge computing technology mean that many controls will be implemented through local devices without having to leave the cloud, and processing will be done at the local edge computing layer. This will undoubtedly greatly improve the processing efficiency and reduce the load on the cloud. As it gets closer to the user, it also provides a faster response to the user and resolves the need at the edge.

Edge calculations emphasize edges. If cloud computing means that all data is aggregated into the back-end data center processing, edge computing is to implement edge intelligence on the edge of the network near the object or data source. Based on this feature, edge computing can achieve high-frequency interaction and real-time transmission of data, so it is expected to shine in the era of Internet of Things and artificial intelligence. Relevant projections show that more than 50 billion terminals will be networked with devices by 2020, and more than 50% of data in the future will need to be analyzed, processed and stored on the edge of the network.

With the development of the Internet of Things and cloud computing, edge computing is on the rise. Edge computing originates from the industrial field and is mainly deployed on terminal equipment or network nodes. It is designed to help equipment in industrial production to have near-end decision-making control when data is not uploaded to the cloud. As the edge calculation heats up, the difference between edge calculation and fog calculation, and how edge calculations are deployed in layers, it becomes a concern of the industry.

If you pay attention to "edge calculation", you will find that it has a brother called "fog calculation". Most published articles have similar interpretations of the term: they are all relative to the "cloud computing", calculated at the edge of the network, closer to the native data (physical perception).

Previously, most of the Internet information processing models we saw were “end-pipe-cloud” models. At the application site, the “end” is only responsible for collecting data and executing instructions, while the “cloud” is responsible for all data analysis and control logic functions. "Edge calculation" or "fog calculation" is to implement part of the data analysis and control logic functions in the vicinity of the application scene. Therefore, there is also a very image called "sticking calculation".

Although the concept of “edge calculation” and “fog calculation” is similar, there are still differences.

"Edge computing" originates from the industrial field and is mainly deployed on terminal devices or network access points. It is now ubiquitous in industrial IoT (Embedded Internet of Things) applications, manufacturing, retail, ATM, smartphones and virtual/mixed reality. Edge computing enables devices in industrial production to have near-end decision control without the help of cloud computing.

"Fog CompuTIng", shelled in "cloud computing", refers to the (partial) function of cloud computing deployed in devices at the edge of the network, localized centralized computing. It is actually an extended concept of cloud computing (Cloud CompuTIng), proposed by Cisco in 2011.

It can be seen that there are still some differences between “edge calculation” and “fog calculation”. The edge calculation is mainly in the "end", which refers to the electronic terminal device or sensor; and the fog calculation is still in the "cloud", deployed in the data centralized site in a certain area. Using a smart home (WiFi) network, for example, an off-net computing performed by an app in a cell phone is edge computing, while a home smart box (smart WiFi gateway) is the subject of fog computing.

Although there are differences between the two, there are currently some articles that do not strictly distinguish between the two. In fact, due to the wide range of IoT business scenarios, the calculations applied to the "end" and "gateway" will be involved. Therefore, since they are all relative to "cloud computing", there is no need to distinguish the location of the deployment (production equipment, sensing devices, gateways/servers), and generally use "edge computing" as a representation.

The difference between "edge calculation" and "fog calculation" gives us inspiration: the computing power in the Internet of Things, with the characteristics of hierarchical deployment. This feature, unlike the cloud computing deployment model in the Internet, can be discussed in two dimensions.

Reference to the edge architecture model of the Internet of Things

Edge Computing Alliance ECC defines four areas for edge computing: device domain (sensing and control layer), network domain (connection and network layer), data domain (storage and service layer), and application domain (service and intelligence layer). These four "layer domains" are the calculation objects of the edge calculation.

Device domain: Edge computing At this level, the perceived information can be directly processed and processed. For example, the ability to directly deploy intelligent authentication in video capture and audio capture; or, like a mobile phone, can be directly converted into text output by voice input.

Network domain: By deploying computing power, automatic conversion of each network protocol is realized, and the data format is standardized. To solve the problem of heterogeneous data in the physical network, it is necessary to deploy edge computing in the network domain to standardize the data format and standardize the data transfer (for example, convert all the perceptual data into MQTT type data and HTTP). transfer). At the same time, the edge calculation of the network domain can also intelligently manage the "converged network", realize network redundancy, ensure network security, and further participate in network optimization.

Data domain: Edge computing makes data management smarter and storage more flexible. First, edge computing can analyze the integrity and consistency of the data, and perform data cleaning to eliminate "dirty" data in the system. Second, edge computing can dynamically deploy compute and storage capabilities as well as system load. Finally, edge computing can maintain efficient coordination with cloud computing and share computing tasks.

Application Domain: Edge Computing provides localized business logic and application intelligence. It enables applications to be flexible and responsive, and to provide localized application services independently when offline (when lost contact with the cloud). Where the IoT is close to users and application scenarios, edge computing is deployed in the above four layers. It enables devices to be intelligently aware, assemble adaptive connection strategies and (digital) deployment strategies, solve data heterogeneity problems in the system, and provide local business logic and even intelligence.

Reference IoT Application / Territory / Coverage

From the original perceptual data to the end of the cloud intelligence, the data will undergo multiple aggregations and calculations according to the needs of the application. For example, from smart homes to smart cities, massive data collection is not achieved in one step. In addition, there are separate applications and services in each phase of data aggregation, which means computing needs for a hierarchical deployment.

Smart cities are divided into four “Internet of Things (size) levels”: homes, communities, communities, cities (see below). Each of these four floors has applications and services, and the scope of services and coverage areas are gradually expanding from home to city. Some of the applications in each level are relatively independent, independent of the upper and lower levels; while some applications are “strategic upgrades”: family doctors (home) community health (community), health care (city).

From the perspective of the Internet of Things level, the relationship between cloud computing and edge computing is differentiated by application:

1. For each service unique to each level, it is only necessary to deploy targeted computing capabilities independently at the corresponding level (only "cloud computing" is required).

2. For penetration (associated) multi-level applications, computing power needs to be deployed from top to bottom. The relationship between the lower layer calculation and the upper layer calculation is the relationship between the edge calculation and the cloud calculation. “Community Medical-Community” is the “cloud” of “family doctor-home” and the “edge” of “health-city”.

3. The relationship between "edge" and "cloud": For a single application, edge computing may be deployed at the upper level (physical network) and cloud computing at the lower level.

It is worth mentioning that an application (such as a community mall) may have the following situation: the core logic and predictive analysis of the application are mainly deployed in the “community” and “community”, and the consumer goods are sold according to the preferences of the regional population; Extract some external data from the “city” level (such as the citywide average price of the goods, etc.); there is not a large amount of application domain computing requirements in the “city” application. If so, then the upper "city" is the "edge" for the lower "community" and "community". Of course, the computing power that the application deploys in the "city" domain is edge computing. The situation of “relationships” may be more in the industrial sector. For example, quality management and process management in industrial production.

Factory quality and process management systems are typically deployed at the production site, and large amounts of production data are stored in “edge” networks. To achieve intelligent production, it is necessary to extract many external information related to quality and supply chain (user complaints, product/parts return information, product life cycle information, partner quality information, etc.). This information will eventually be integrated into the “edge” quality and process management system for quality analysis and forecasting along with the Internet of Things. Obviously, for quality and process management systems, the Internet and the Internet of Things outside of itself are edge networks.

It is foreseeable that the "cloud computing" of industrial production will be deployed more on the edge of the Internet of Things, near the industrial production site. Depending on the needs of the application, computing power is deployed in the various (size) levels of the Internet of Things. Regardless of the level at which "computation" is deployed, if it assumes the primary responsibility of on-site command, it is edge computing; if it assumes the primary responsibility for big data and intelligent forecasting, it belongs to cloud computing.

As applications become more flexible in computing deployments, cloud computing and edge computing will move toward convergence and become increasingly difficult to differentiate. When the Internet of Things is filled with general-purpose computing capabilities that are ubiquitous and ubiquitous, “ubiquitous computing” will emerge. Edge intelligence is the future

Deploying simple application logic at the edge of the Internet of Things cannot meet the needs of multi-modal IoT applications. In the vicinity of the application scenario, certain intelligence must be deployed to build a robust application ecosystem on the edge of the Internet of Things.

The essence of edge computing is cloud computing. The most important ability of edge computing is to inherit the intelligence of cloud computing. As far as the current technological development trend is concerned, it is theoretically possible to do this. For a certain application, after learning enough application scenarios, the neural network algorithm (cloud computing) can be “slimmed down” (reduced) and then deployed at the edge of the network (deploying intelligent edge computing) to form Edge intelligence. In this way, edge intelligence can achieve most of the intelligence of the application scenario even without the support of cloud computing.

For example, on May 23, 2017, AI AlphaGo defeated Ke Jie by 1/4. It is worth noting that the day it was played by the "one" stand-alone version of AlphaGo.

When edge computing becomes edge intelligence, local and edge IoT systems can have autonomous self-discipline behavior. Self-sufficient computing power and intelligence will enable IoT applications to operate relatively independently from "cloud computing."

Edge computing has the characteristics of a hierarchical ("layer domain" and "hierarchy") deployment.

On the one hand, edge computing is deployed on various layers of the edge architecture model. The computing power is deployed in the hierarchical domain at the edge of the physical network, so that the application in the Internet of Things (such as smart home) can also form an information loop of ''perception'-'connection'-'analysis and prediction'-'control'" . Thereby, the information value of various types of data is released.

On the other hand, by deploying computing power in different areas (sizes and sizes) of the Internet of Things, developers can not only build information loops of appropriate size based on business needs and characteristics, but also enable "vertical" services to "loop" between levels. Intertwined, mutual service and value interoperability.

The higher value of edge computing is edge intelligence.

Edge computing is the deployment of intelligent cloud computing. The application realizes the information looping in the Internet of Things, and can realize the intelligence of the whole layer domain such as information decision, behavior feedback, automatic networking and load balancing through edge calculation. In the case of leaving the cloud computing, the application can also operate independently and flexibly, so as to form the "ecology" of the Internet of Things (the mechanism for forming information mutual assistance services between various types of devices) within a small scope of the application scenario.

Disposable Vape

The utility model provides a disposable electronic cigarette, comprising: a hollow shell, the bottom of the shell is provided with a lower cover; the shell contains an atomizer, and the outer side of the atomizer is sheathed with a disposable cigarette A bomb, a microphone cover is arranged under the atomizer, a microphone is covered under the microphone cover, a battery is arranged on one side of the atomizer, and an upper cover is arranged on the top of the casing; The atomizer includes an atomizing corThe utility model provides a disposable electronic cigarette, comprising: a hollow shell, the bottom of the shell is provided with a lower cover; the shell contains an atomizer, and the outer side of the atomizer is sheathed with a disposable cigarette A bomb, a microphone cover is arranged under the atomizer, a microphone is covered under the microphone cover, a battery is arranged on one side of the atomizer, and an upper cover is arranged on the top of the casing; The atomizer includes an atomizing core, an oil-absorbing cotton sleeved on the outside of the atomizing core, and an atomizer outer tube sleeved on the outside of the oil-absorbing cotton. The disposable electronic cigarette provided by the utility model absorbs the smoke oil on the surface through the absorbing cotton, and then atomizes the smoke through the atomizing core, which greatly reduces the risk of oil leakage, at the same time, reduces the burning of cotton and ensures the smoking taste.e, an oil-absorbing cotton sleeved on the outside of the atomizing core, and an atomizer outer tube sleeved on the outside of the oil-absorbing cotton. The disposable electronic cigarette provided by the utility model absorbs the smoke oil on the surface through the absorbing cotton, and then atomizes the smoke through the atomizing core, which greatly reduces the risk of oil leakage, at the same time, reduces the burning of cotton and ensures the smoking taste.

disposable vape,vape pen,vape accessories,vape store,vape battery,e-cigarettes,Disposable Pod

Suizhou simi intelligent technology development co., LTD , https://www.msmvape.com