Edge computing: trillions of blue ocean market accelerated outbreak
recently, China Unicom's first MEC scale commercial network and ecological cooperation conference was successfully held in the cloud, and officially released the world's first MEC scale commercial network and the 5g MEC edge cloud platform architecture and commercial practice white paper
trendforce's latest forecast shows that the market for edge computing products and services will grow at a CAGR of more than 30% from 2018 to 2022. Among them, the Asia Pacific market is developing the fastest, and the compound annual growth rate is expected to reach 46.7%. According to Cisco's prediction, one third of the data traffic will completely bypass the center and sink to the metropolitan area, closer to the edge
with the gradual promotion of 5g commerce, marginal computing (MEC), which can connect 5g commercial potential and downstream industry opportunities, is moving towards the gale, showing strong market demand in the fields of industrial interconnection, autonomous driving, IOT and so on. IDC predicts that by 2020, more than 50billion terminals will be connected with devices, and more than 50% of the data needs to be analyzed, processed and stored at the edge of the network. Therefore, the size of the marginal computing market will exceed trillion yuan
the cloud computing industry has developed rapidly in the past decade, and its centralized data processing architecture has been recognized by the market. However, with the deepening of industrial interconnection and the popularity of IOT technology in the 5g era, more and more devices, including cameras, sensors, etc., are connected to the cloud, and real-time processing of massive device data will also pose a certain pressure on the cloud centric model
in this case, edge computing can just play a complementary role to cloud services, and optimize the cost pressure, single point availability and other issues brought about by the deployment of centralized cloud computing center to a certain extent. In short, edge cloud computing services combine traditional cloud computing with edge computing to form a cloud edge end collaborative computing architecture
the core of edge computing is to migrate computing tasks from the cloud computing center to the edge devices that generate source data. Compared with traditional cloud computing, it has higher security, lower latency, and reduced bandwidth costs
edge computing is very important for 5g applications such as AR, cloud games, low latency HD video streaming, IOT and autopilot, which can effectively reduce latency and network congestion. Edge computing allows thousands of IOT device data to be processed locally, which can significantly reduce cloud storage and transmission costs, while improving privacy and security
from the perspective of 5g construction in South Korea, Korean operators SKT and KT will lead the construction of global edge computing in 2019, and their main application direction is the delay sensitive multi player game business. In the future, edge computing is expected to be widely used in vertical industries such as industry, automobile, energy, medical treatment and retail
cloud computing services support data intensive applications. As a useful supplement to cloud computing, edge computing can use edge node processing and analysis to shorten the response time of devices and reduce the data flow from devices to cloud data centers, so as to allocate resources more effectively in the network
calculate from the edge and say to soft he: "The central government requires the supply side reform to add and subtract components and hardware. In terms of the requirements, the software platform needs to consider the introduction of cloud concept, Cloud Architecture, cloud technology, and provide end-to-end real-time, collaborative intelligence, reliable, dynamic reset and other capabilities. The hardware platform needs to consider heterogeneous computing capabilities, such as Kunpeng, arm, x86, GPU, NPU, FPGA, etc.
with the development of IOT technology, more real-time, short-cycle data analysis needs The increase and the increase in the requirements for the timely processing and implementation of local businesses at the provincial (District, city) level will drive the promotion of the application of edge computing, and with the increase in the demand for edge computing servers
the integration of edge computing, cloud computing and 5g supports massive machine communication, promotes the deep integration of typical application scenarios represented by smart cities and smart homes with mobile communication, and accelerates the application and landing of IOT and related technologies in vertical industries
landfill and other treatment can be carried out.
at present, the main players in the global edge computing ecosystem include hardware manufacturers, software manufacturers, scene service providers, operators, cloud computing manufacturers and CDN manufacturers. Hardware and chip enterprises have successively launched edge AI chips, which are applied in a large number of scenarios such as intelligent transportation, vr/ar, car service, intelligent manufacturing, etc; With the accumulated advantages of the cloud era, cloud computing giants have extended their tentacles to the edge computing era; Established CDN enterprises are using their traditional CDN nodes, But Ray Gibbs explained: "The improvement in material processing has begun to pay off for edge computing services; operators have begun to deploy MEC, integrate the industry macroscopically from multiple perspectives, and promote the establishment of global unified standards.
in addition to traditional communication equipment suppliers such as Huawei, Ericsson, Nokia, Samsung, cloud computing manufacturers Amazon, Google, Akamai, etc. are also actively entering to integrate their computing resources Sink into the field of edge computing. At present, there are many start-ups providing MEC as service edge computing services abroad, including packet, vapor IO, edge micro, edgeconnex, etc. 5g edge computing will complement cloud computing and is expected to gradually cultivate a new business ecosystem
according to the edge computing alliance industry white paper 3.0 of the edge computing industry alliance, edge computing is divided into three layers: terminal equipment, edge and cloud, and inter layer and cross layer communication can be carried out between each layer
the terminal layer is composed of various devices (PC, intelligent, etc.), which mainly completes the function of collecting raw data and reporting, and takes the form of event source as the input of application services
the edge computing layer is composed of network edge nodes, which are widely distributed between terminal devices and computing centers. The computing and storage resources of edge nodes are very different, and the resources of edge nodes are dynamic. The edge computing layer realizes the basic service response by reasonably deploying and allocating the computing and storage capacity on the edge of the network
among the joint services of cloud computing, cloud computing is still the most powerful data processing center. The reported data of the edge computing layer will be permanently stored in the cloud computing center, and the analysis tasks that cannot be processed by the edge computing layer and the processing tasks of comprehensive global information still need to be completed in the cloud computing center
5g era, the network architecture will be further upgraded. With data center as the infrastructure and cloud as the basic platform, four cloud centers will be formed, namely, access cloud, edge cloud, convergence cloud and core cloud. Under the edge computing technology, the 5g network introduces new nodes of UPF and PCF, through which the diversion function and diversion strategy control are realized respectively, so as to solve the problems of billing and strategy, and promote the real commercialization of edge computing
among many emerging businesses in vertical industries, the demand for edge computing is mainly reflected in three aspects: time delay, bandwidth and security. At present, the demand for edge computing is the most clear in the four vertical fields of intelligent manufacturing, smart city, live game and car service. IDC predicts that by 2020, more than 50billion terminals will be connected to devices, and 50% of physical connections will face network bandwidth constraints. With the advent of the era of Pan IOT, data will grow exponentially, and edge computing is expected to usher in a rapid outbreak
Copyright © 2011 JIN SHI