撰文/國際商務部
編輯/品牌推广部
With thrive development of Internet of things (IoT), we are in a new networked systems era. The total amount of data is unprecedently high and most of them are generated at the edge of the network due to the increase quantity of the embedded devices.
隨著物聯網的蓬勃發展，我們正處在一個新的網絡系統時代。由於嵌入式設備數量的增加，數據總量空前的高，且大部分都是在網絡邊緣產生的。
Edge computing is an open platform, which is physically close to the edge of the data source, integrates the core capabilities of network, computing, storage and application, and provides the computing mode of edge intelligent services nearby. This advanced distributed computing paradigm places the processing of data, the operation of application programs, and even the realization of some functional services from the network center to the nodes on the edge of the network, and processes the data nearby without uploading a large amount of data to the remote core management platform, so as to provide a more efficient and secure computing paradigm to cope with the challenges such as reliability, safety and timeliness in the new society, environment and industrial applications.
邊緣計算是一個開放的平臺，它在物理上接近數據源的邊緣，集網絡、計算、存儲、應用等核心功能於一體，提供了附近邊緣智能服務的計算模式。這種先進的分布式計算範式將數據資料的處理、應用程式的運行甚至一些功能服務的實現，由網絡中心下放到網絡邊緣的節點上，就近處理數據，而不需要將大量數據上傳到遠端的核心管理平臺，從而提供一種更加高效、安全的計算範式來以此應對來自新社會、環境、工業應用中可靠性、安全性、及時性的挑戰。Applying edge computing in the field of Internet of things (IoT), physical computing and virtual and augmented reality (VR/AR), has been attracting much attention in recent years.近年來，邊緣計算在物聯網、物理計算、虛擬增強現實等領域的應用受到了廣泛的關注。
There are three types of edge scenarios to consider when deploying edge services: personal edge, business edge and cloud edge.
在部署邊緣服務時，需要考慮三種類型的邊緣場景：個人邊緣、業務邊緣和雲邊緣。
Personal edge computing revolves around everyday life, such as smart phones, home robots, smart glasses, medical sensors, wearing watches, smart speakers, and other home automation systems.Personal edge devices are generally mobile, so they are also called mobile edge computing (MEC).個人邊緣計算在日常生活中隨處可見，如智慧型手機、家庭機器人、智能眼鏡、醫療傳感器、佩戴手錶、智能揚聲器和其他家庭自動化系統。個人邊緣設備通常是移動的，因此也被稱為移動邊緣計算（MEC）。Business edge is used to gather the information of personal edge devices, where robot, sensor and other information are collected and processed. Such devices can be deployed in office area or home area to support information concentration, interaction and processing within the area.業務邊緣用於收集個人邊緣設備的信息，用於收集和處理機器人、傳感器等信息。這些設備可以部署在辦公區域或家庭區域，以支持區域內的信息集中、交互和處理。Cloud edge is a complex IoT application involving the collaboration of multiple cloud platforms. The rise of voice processing, face recognition and other vertical cloud platforms has improved the intelligence of the IoT, but also put forward higher requirements for the cooperation between platforms. Cloud edge provides data analysis, data interaction and data collaboration functions on different cloud platform sides.雲邊緣是一個複雜的物聯網應用，涉及多個雲平臺的協作。語音處理、人臉識別等垂直雲平臺的興起，提高了物聯網的智能化水平，也對平臺間的合作提出了更高的要求。雲邊緣在不同的雲平臺端提供數據分析、數據交互和數據協作功能。
Benefits of edge computing1. Reduce latency and assist in real-time decision makingBenefits of edge computing include the ability to conduct on-site massive data analysis and aggregation, which allows for near real-time decision making.邊緣計算的好處包括能夠進行現場海量數據分析和聚合，從而實現近乎實時的決策。One of the real world examples of this advantage is in the use of virtual and augmented reality. Virtual and augmented reality often suffer from insufficient bandwidth and high latency. It’s common for folks using these technologies to experience sickness or lags in the computing power that break the immersiveness of the experience.這種優勢在現實世界中的一個例子就是虛擬實境和增強現實的使用。虛擬實境和增強現實經常受到帶寬不足和高延遲的困擾。對於使用這些技術的人來說，經常會遇到不流暢或計算能力落後，從而破壞了體驗的沉浸感。Edge computing allows the compute-intensive parts of the rendering pipeline to be offloaded to the cloud, preventing these problems from occurring. These same processes are implemented in the Internet of Things (IoT), like smart cars, so that users can have their cars (and other IoT devices) process data close to the terminal end and make decisions as close to real time as possible.邊緣計算允許將渲染管道中計算密集的部分卸載到雲，從而防止這些問題的發生。這些步驟也應用在物聯網（IoT）中，比如智能汽車，這樣用戶就可以使他們的汽車（和其他IoT設備）在靠近終端的地方處理數據，從而儘可能接近實時地做出決策。2. Enhance data security managementEdge computing further reduces the risk of exposing sensitive data by keeping all of that computing power local, thereby allowing companies to better control the proliferation of information (like industry trade secrets) or meet regulatory policies (like the GDPR).邊緣計算通過保持所有計算能力的本地性，進一步降低了暴露敏感數據的風險，從而允許公司更好地控制信息（如行業商業機密）的擴散或滿足監管政策（如GDPR）。3. Reduce bandwith transmission pressureFinally, enterprise customers benefit from the resiliency and costs associated with edge computing. By keeping computing power local, regional sites can continue to operate independently from a core site even if something causes core site to stop operating. The consumption of paying for bandwidth to take your data back and forth between your core and regional sites is also greatly reduced by keeping that compute processing power closer to its source.最後，企業客戶可從與邊緣計算相關的彈性和成本中獲益。通過保持本地計算能力，區域站點可以繼續獨立於核心站點運行，即使某些因素導致核心站點停止運行。在核心站點和區域站點之間來回傳輸數據所支付的帶寬開銷也可以通過使計算處理能力更接近其來源而大大降低。Challenges of edge computingEdge computing is mainly a problem of highly-distributed scale:Scaling out to many small sites can be more complicated than adding the equivalent capacity to a single core datacenter. The increased overhead of physical locations can be difficult for smaller companies to manage.擴展到許多小型站點可能比將等效容量添加到單個核心數據中心更為複雜。對於較小的公司來說，增加的物理位置開銷可能很難管理。Edge computing sites are usually remote with limited or no on-site technical expertise. If something fails on-site, you need to have an infrastructure in place that can be fixed easily by non-technical local labor and further managed centrally by a small number of experts located elsewhere. 邊緣計算站點通常是遠程的，具備有限或無現場技術專業知識。如果現場出現故障，則需要有一個基礎設施，可以由非技術性本地勞動力輕鬆修復，且該設施可由其他地方的少數專家進一步集中管理。The storage capacity of edge nodes is limited, so how to deal with these data and how to save them become the key to the problem. If too much original data is filtered out, it will lead to unreliable data report of edge nodes. If a large amount of original data is retained, the storage of edge nodes will be a new problem.邊緣節點的存儲容量是有限的，如何處理和保存這些數據就成了問題的關鍵。如果過濾掉太多的原始數據，會導致邊緣節點的數據報告不可靠。如果保留大量原始數據，邊緣節點的存儲將成為一個新的問題。By 2020, there will be 50 billion terminals and devices connected to the Internet. How to find resources and services in the distributed computing environment and allocate tasks effectively is an area to be expanded. In order to make full use of the edge devices of the network, it is necessary to establish some discovery mechanism (such as naming mechanism, network protocol, etc.), find the appropriate nodes that can be deployed in a decentralized way, and realize the dynamic and large-scale deployment of computing and storage capabilities, as well as the efficient collaboration and seamless docking between the cloud and the device.到2020年將有500億的終端和設備聯網，如何在分布式計算環境中發現資源和服務，並有效的分配任務是一個有待拓展的領域。為了充分利用網絡的邊緣設備，需要建立某種發現機制（如命名機制，網絡協議等），找到可以分散式部署的適當節點，實現動態、大規模地部署運算和存儲能力以及雲端和設備端的高效協同、無縫對接。Opportunities for edge comouting market1. Architecture and languageAs the number of edge nodes supporting general-purpose computing increases, so will the need for development frameworks and toolkits.隨著支持通用計算的邊緣節點數量的增加，開發框架和工具包的需求也會增加。2. Lightweight libraries and algorithmsDue to hardware limitations, edge nodes do not support large-scale software. Edge analysis requires lightweight algorithms, which can perform reasonable machine learning or data processing tasks.由於硬體限制，邊緣節點不支持大型軟體。邊緣分析需要輕量級算法，這些算法可以執行合理的機器學習或數據處理任務。3. Micro operating system and Virtualization
Research Based on micro operating system or micro kernel and maturity of container technology (such as docker) can solve the challenges of application deployment on heterogeneous edge nodes.
基於微作業系統或微內核的研究和成熟的容器技術（如docker）可以解決異構邊緣節點上應用部署的難題。
2019蒼穹數碼政府信息化系列論壇
不動產登記創新服務論壇
正在火熱報名中！
11月30日-合肥市-即將啟幕
報名截止日期11月28日！
相約合肥，我們不見不散！
參會報名，點擊這裡
↓↓↓