From CDN to edge computing, which one is closer to the water?

Since the birth of CDN, there have been three generations of traditional CDN, cloud CDN and shared CDN, but none of them have been able to jump out of the technical scope of traditional CDN and have not achieved substantial cost advantages. Although edge computing is an emerging technology, many corporate giants, such as Cisco, Intel, Huawei, and ZTE, have already joined it. The concept of CDN and edge computing is to process data as close to the edge where data is generated as possible. Compared with other technologies, can the transformation from CDN to edge computing achieve the first-mover advantage?

What is a CDN?

CDN stands for Content Distribution Network, and its purpose is to enable users to obtain the required content nearby, solve the problem of Internet network congestion, and improve the response speed of users visiting websites. CDN is a network of edge servers that provides optimized distribution or delivery of online content. A large number of edge servers work together to transmit data through a private global backbone network, bypassing most of the congested public and Internet Service Provider (ISP) networks, thereby improving the speed and efficiency of content delivery. The main goal of CDN is to improve Web performance by reducing the time required to transmit content and rich media to users' Internet devices.

Any content on the Internet can be delivered through a CDN. For example, images, style sheets, JavaScript files, file downloads, live broadcasts, games, and so on. In fact, most of the Internet is already delivered through CDNs, and any device connected to the Internet interacts with a CDN, because CDNs are not only used to deliver content on computers, but also on mobile devices, smart TVs, set-top boxes, and many other connected devices that require fast, reliable, and secure online data transmission.

What is edge computing?

Edge computing refers to providing the nearest service on the side close to the source of objects or data, using an open platform that integrates network, computing, storage, and application core capabilities. Edge computing is a distributed computing architecture with decentralized processing capabilities that supports mobile computing and Internet of Things (IoT) technology. Under this architecture, edge computing can process or store key data locally and push all received data to a central data center or cloud storage repository. Edge computing classifies data locally and processes part of the data locally, thereby reducing the return traffic to the central storage repository, which can speed up the processing and transmission of data and reduce latency.

At present, the most popular use case of edge computing is autonomous driving, because edge computing can accelerate data flow, including real-time data processing without delay. Due to the particularity of autonomous driving, we cannot send its numerous sensors to the cloud and wait for responses, and the existence of delays is completely unacceptable. The use cases of edge computing also include CDN, Internet of Things, smart security, and blockchain.

The connection between CDN and edge computing

The common point between edge computing and CDN is that they require transmission capabilities as close as possible to where the data is generated. CDN needs to transmit and store a large amount of content data, and edge computing happens to provide storage services. Therefore, the deployment methods of the two are similar, both are close to the edge of the network, and bandwidth lines can be reused.

The main computing nodes and applications of edge computing are distributed and deployed in data centers close to the terminals, which makes the service response performance and reliability higher than the traditional centralized cloud computing concept. The CDN nodes can be fully reused to provide computing services.

The difference between CDN and edge computing

The edge in the CDN concept is to improve the node's ability to transmit data by caching data, with the focus on transmission capacity. Edge computing actually uses the edge area close to the data source to perform data computing and classification, with the focus on computing power.

It can be seen that the main ideas of CDN and edge computing are different. CDN will still trace the data back to the data center for processing, which is still very costly. However, edge computing does not need to transmit every piece of data to the cloud. It only needs to use the devices at the edge of the data to perform data calculations and processing, which can reduce the data flow from the device to the cloud. Therefore, edge computing can reduce the pressure of data explosion and network traffic. When transmitting to the cloud, some simple data processing is performed through edge nodes, thereby reducing device response time and latency.

From CDN to Edge Computing

The transition from CDN to edge computing is imperative. Due to the rapid development of high-definition video, VR/AR, big data, the Internet of Things, and artificial intelligence, the traditional CDN method of tracing data back to the central cloud is too costly and difficult to meet the increasing storage, computing, and interaction needs of massive data. Therefore, CDN must evolve from a traditional IO-intensive system centered on cache services to an edge computing system and build a content computing network to solve the connection challenges brought by the future Internet of Things. In this process, the CDN industry will also complete a second iteration.

The future Internet of Things, AR/VR scenarios, big data and artificial intelligence industries actually have a strong demand for near-field computing. Edge computing can ensure that a large amount of computing needs are completed in an area very close to the terminal and achieve low-latency services.

Enterprises expanding from CDN to edge computing

CDN's inherent edge node attributes give it a first-mover advantage in the edge computing market. CDN itself is the prototype of edge computing. From this perspective, it is easier to expand from CDN to edge computing in terms of technology implementation, and of course everyone does this.

• Cisco and Reliance Jio are now jointly leveraging the power of multi-access edge computing, starting with a use case to further optimize and enhance the online video experience by developing a mobile content delivery network (CDN). In this concept, the CDN is integrated into the mobile LTE network, and with the help of mobile CDN, mobile operators can deliver content through edge cloudlets, thereby providing a better user experience with lower latency and higher performance.
• Ericsson said that before the rise of edge computing, CDN has always been the representative of edge networks, but the main purpose of CDN is local storage, not local computing. Now, Ericsson hopes to combine the two. At the Mobile World Congress, Richard Dawson, director of CDN business development at Ericsson, demonstrated the company's unified distribution network solution.
• Alibaba Cloud believes that the central computing of traditional cloud computing must be combined with edge computing. Through the "cloud + edge + terminal" trinity computing model, a more real-time response can be achieved. In March this year, Alibaba Cloud launched Link Edge, the first IoT edge computing product. Its second edge computing product is the edge node service ENS. At present, Alibaba Cloud CDN has more than 1,300 nodes, covering more than 70 countries and regions, and has a bandwidth capacity of nearly 90T.
• Tencent Cloud has not yet launched an open framework based on the edge, but it said that it will design an overall CDN framework in the future to sink CDN nodes to the edge, closer to users. For Tencent Cloud, the future CDN system will also include AI + edge computing + Internet of Things.
• Wangsu Technology is pushing for edge computing, which is a necessary condition for CDN network intelligence. As early as 2016, Wangsu Technology was the first in the industry to propose that the next development direction of CDN is edge computing. At present, Wangsu Technology will gradually launch edge platform services. In the area of ​​microservices, Wangsu Technology can already support customers to implement them on the edge computing platform.

Summarize

In general, CDNs in the future will need a large number of edge devices. Whether it is shifting from CDN to edge computing, or adding the concept of edge computing to the original CDN system, using edge computing to enhance CDN's own competitiveness is a good choice. Edge computing can help CDN become smarter, more efficient and more stable. Edge computing has a bright future. It is expected that more than 50% of data will need to be analyzed, processed and stored at the edge of the network in the future, which requires edge computing to play a big role.