Edge computing fever soars no longer stops at concepts
There is no doubt that cloud-based AI computing has become mainstream in the current AI landscape. For the cutting edge technology industry, it's just always going to create all kinds of windfalls for everyone. In the big data and cloud computing industry, it's clear that one word has been soaring at a particularly fast pace lately: edge computing.
Edge computing is more than a concept Beginning to take the lead in industry applications
In fact, edge computing is not a new thing; its computing principles have been proposed many years ago. But with the recent development of the IoT industry and the arrival of the 5G era, edge computing, which is closely related to these areas, has been thrust back into the limelight. Everything is connected, edge first. The BI Intelligence report predicts that edge computing will be first to be used in the manufacturing, utilities, energy and transportation industries.
Edge computing is an open platform that incorporates core capabilities of networking, computing, storage, and applications on the edge side of the network close to the source of things or data. Edge computing and cloud computing work in concert with each other to help digital transformation across industries. It provides intelligent interconnected services close to home to meet the industry's critical needs for business real time, business intelligence, data aggregation and interoperability, security and privacy protection during digital transformation.
By 2020, each person will generate 1.7MB of data per second and 237 million IoT wearable devices will be shipped, according to a study by the International Telecommunication Union's Telecommunication Standards Branch, ITU-T. IDC has also released predictions that by 2018, 50% of IoT networks will face network bandwidth limitations and 40% of data will need to be analyzed, processed and stored at the edge side of the network, and by 2025, that number will exceed 50%.
Edge Computing Technology Explained and Relationship to Fog Computing
When talking about edge computing, there is no getting around a concept called "fog computing". As soon as you hear the name you know this is a variation from the cloud one. The principle of fog computing is the same as cloud computing in that data is uploaded to a remote center for analysis, storage and processing. However, in contrast to cloud computing, where all data is transported to a centralized hub, the fog computing model has many decentralized central nodes, known as "fog nodes", to handle the process. This enables faster arithmetic processing and more efficient arithmetic results.
If cloud computing is about sending everything to the clouds in the sky, fog computing is about sending data to the fog around us, a logic known as "decentralized cloud computing".
But the idea of decentralizing cloud computing has produced more than just one result, fog computing. One more radical idea is that if we're going to spread out, then why not simply complete the calculation at the closest distance? Isn't running to the cloud or running to a decentralized center just running? And so there is what is called edge computing.
The operation can be done within a large computing device, or within a medium or small computing device, or a local end network. And the devices used for edge computing can be both mobile devices like smartphones, home terminals like PCs and smart homes, and even municipal terminals like ATMs and cameras. In short, in a word, count wherever you can near you, the closer the better. To give an inappropriate but very easy to understand example: if you have a file in your phone that is too big to fit, the cloud computing solution is to send it to a server center thousands of miles away to help you store it, fog computing is to send it to a server in your cell to help you store it, edge computing is to see who has space on your computer, refrigerator, or washing machine to help you store it ......
Obviously, the closest distance between edge computing computing computing devices and end devices, the most direct transmission effect, can also make use of proximity networks and local area networks, and of course the computing efficiency will be a little higher.
Technology comparison between edge computing and cloud computing
Currently, cloud computing is a big trend in the industry, and for cloud computing, all data is aggregated to the back-end data center for completion. Among the three roles of "cloud, management and end", cloud computing is more focused on the "cloud", which is the place to achieve the final data analysis and application.
However, in edge computing, the emphasis is on the physical area where the "edge" or "end" is located. In this area, if network, computing, storage and other resources can be provided for the "end" nearby, it is obvious that real-time and other business needs can be easily met, which is the biggest difference between "edge computing" and "cloud computing". "Edge computing" also places special emphasis on the role of computing.
Cloud computing is the centralization of computing services in its simplest form, leveraging shared data center infrastructure and economies of scale to reduce costs. However, due to the number of router hops, packet latency from the introduction of virtualization technologies or server latency within the data center is a key issue for cloud migration, the concept of edge computing has been gaining steam in recent years, while edge computing is also a driver of innovation in OpenStack.
Edge computing, a decentralized computing architecture, moves the computing of applications, data and services from the central node of the network to the edge nodes on the network logic to process them. Edge computing breaks up large services that would otherwise be handled entirely by a central node, cutting them into smaller and more manageable parts, and distributing them to edge nodes for processing. Edge nodes are closer to user end devices, which can speed up the processing and delivery of data and reduce latency.
Summary. Today, the rise to fame of edge computing in a short period of time is inextricably linked to the development of several areas of algorithms, data and transmission networks. In the age of the Internet of Everything, edge computing covers a very wide range of technologies and is now present in almost every industry. In the past six months, a large number of domestic smart home and IoT companies with edge computing as their selling point have surged forward, hoping to become "pigs on the wind" with the explosion of edge computing. Edge computing isn't just a concept anymore, it's shining into reality.
(This article is based on Titanium Media, Huawei Agile Network, IoT Think Tank, Deep Technology, China Security Industry Network, etc., and edited and published by China Security Exhibition Network)