A detailed table of contents of CDN technology and a brief analysis of CDN implementation principles

Due to the performance bottleneck of the user's access to the origin server, CDN technology is used to cache the content of the origin server to multiple nodes. When a user sends a request to the domain name of the origin server, the request is scheduled to the service node closest to the user, and the service node directly responds quickly, effectively reducing user access latency and improving availability.

To understand CDN, you need to understand the knowledge points first:

No matter how much it is, it's not as good as practical application, let's take an example to understand the DNS resolution process.

nslookup is the command for dns resolution, and server is the ip of the company's internal dns serveris the IP address of the requested domain name.

The detailed process can be viewed using the dig command.

Let's take a step-by-step analysis of the parsing process

cname stands for: alias record. This type of record allows you to map multiple names to another domain name.

It is usually used on computers that provide both www and mail services. For example, there is a computer named "a record"). It provides both www and mail services for the convenience of users.

You can set two aliases (cnames) for this computer: www and mail. The full name of these two aliases.

and " " actually they all point to " ".

The implementation principle in CDN is to add CNMAE to the domain name of the origin server, alias the domain name of the acceleration node. When a user makes a request to the origin server, the DNS server finds that there is a cnmae record when the origin server resolves the domain name of the origin server, and the DNS server sends a request to the CNAME domain name, and the request is scheduled to the domain name of the acceleration node.

For example, if the domain name of your business origin server is , and your users initiate an HTTP request after the domain name is connected to the CDN and you start using the acceleration service, the actual processing process is shown in the following figure: link.

CDN is a content distribution network built on the network, relying on edge servers deployed in various places, through the load balancing, content distribution, scheduling and other functional modules of the central platform, so that users can obtain the content they need nearby, reduce network congestion, and improve user access response speed and hit rate. The key technologies of CDN mainly include content storage and distribution technology.

To put it bluntly, the core technologies and functions are: content distribution and caching;

The best way is to buy a secondary server in Beijing, and copy the content of Guangzhou to the secondary server, which is called CDN server.

The content of "Detailed Explanation of CDN Technology" includes the development history, key technologies, commercial service status of CDN technology, and the future development prospects, focusing on the key functional modules of the CDN system, GSLB, SLB, and Cache, in addition to the technical principles, it also explains some protocols and development tools involved in the implementation of these functional modules, hoping to help readers understand the technology of CDN, and have some preliminary experience in the design and development of CDN system.

Detailed Explanation of CDN Technology is suitable for professionals engaged in Internet development and operation, telecom operation service practitioners, and college students in related majors.

A CDN is a distributed content distribution network built on a data network. The function of CDN is to use the streaming server cluster technology to overcome the shortcomings of the output bandwidth and concurrency capacity of the stand-alone system, which can greatly increase the number of concurrent streams supported by the system and reduce or avoid the adverse effects caused by single point failure.

CDN uses global load balancing technology to direct user access to the nearest working streaming server, which responds directly to the user's request. If there is no content that the user wants to access in the defense server, the corresponding content will be automatically crawled from the original server according to the configuration and provided to the user. IPTV can use CDN to provide users with VOD services, and after distributing ** content to CDN nodes close to the user through CDN, the service quality of the end-to-end key office can be ensured to a certain extent.

CDN is a combination of technologies that includes several important parts such as the origin server, cache server, intelligent DNS, and client. The client or user side is the ordinary user who initiates the access, and the general access method is the browser. I won't go into detail here.

The main technical means to implement CDN are cache and mirror server. It can work in two ways: DNS resolution or HTTP redirection, and complete the delivery and synchronous update of content through the cache server or a remote mirror site. The accuracy rate of user location judgment in DNS mode is greater than 85%, and the accuracy rate of HTTP mode is more than 99%. In general, the ratio of the amount of incoming data accessed by users in each cache server cluster to the amount of data accessed by the cache server to the original content is 2:

Between 1 and 3:1, that is, 50% to 70% of the amount of duplicate access data to the original **; For mirroring, except for the traffic of data synchronization, the rest is completed locally and does not access the original server.

Mirror site servers are often seen as they allow content to be distributed straightforwardly and are suitable for static and quasi-dynamic data synchronization. However, the cost of purchasing and maintaining a new server is high, and it is also necessary to set up mirror servers in various regions and have professional technicians to manage and maintain them. Large** servers are updated at any time, and the demand for bandwidth will also increase significantly, so the average Internet company will not build too many mirror servers.

Caching means are less expensive and suitable for static content. Internet statistics show that more than 80% of users often access 20% of the content, under this law, the cache server can handle most of the customer's static requests, and the original www server only needs to process about 20% of the non-cached requests and dynamic requests, so it greatly speeds up the response time of customer requests, and reduces the load of the original www server. According to a survey by IDC in the United States, the caching market, which is an important indicator of CDN, is growing at a rate of nearly 100% per year, and the global turnover will reach $4.5 billion in 2004.

The development of network streaming** will also stimulate the demand of this market.