Architecture analysis of Taobao (group)

The system architecture of a mature large website (such as Taobao, JD.com, etc.) needs to consider many complex factors, because the data volume of a large website like Taobao is much larger than that of an ordinary website, so the design architecture is also much more complex, considering both cost factors and access speed and security. Here I will simply analyze the system architecture of Taobao's website.

As a large shopping website, Taobao has a large amount of data, so unlike ordinary websites, Taobao needs to use various methods to ensure the normal operation of the server and the good experience of users when purchasing. Mainly in the following ways:

1. Separation of applications, data, and files 2. Using cache to improve website performance 3. Using CDN and reverse proxy to improve access speed 4. Using distributed file system

5. Splitting the application server for business

First of all, as a large shopping website, one server has definitely met the performance requirements, so the application, database, and files are deployed on separate servers, and different hardware is configured according to the purpose of the server to achieve the best performance effect. While optimizing performance through hardware, we also optimize performance through software. In most website systems, cache technology is used to improve system performance. The use of cache is mainly due to the existence of hot data. Most website access follows the 28 principle (that is, 80% of access requests eventually fall on 20% of the data), so we can cache hot data, reduce the access path of these data, and improve user experience.

Due to the complexity of functions and the long user access path, Taobao chooses to cache these data to improve user access speed. Common ways to implement cache are local cache and distributed cache. Local cache, as the name suggests, caches data locally on the application server. It can exist in memory or in files. The characteristic of local cache is fast speed, but because the local space is limited, the amount of cached data is also limited. The characteristic of distributed cache is that it can cache massive amounts of data and is very easy to expand. It is often used in portal websites, but the speed is not as fast as local cache. At the same time, a load balancing server is provided to share the pressure of the main server.

Use CDN and reverse proxy to improve website performance. Since Taobao's servers cannot be distributed everywhere in China, users in different regions need to access the server through interconnected routers through paths of different lengths, and the return path is the same, so the data transmission time is relatively long. For this situation, CDN is often used to solve it. CDN caches the data content in the operator's computer room. When users access it, they first obtain data from the nearest operator, which greatly reduces the network access path.

Here is a brief introduction to the principle of CDN. The full name of CDN is Content Delivery Network. CDN is a strategically deployed overall system that technically solves the fundamental reason for the slow response speed of users accessing websites due to small network bandwidth, large user visits, and uneven distribution of outlets. The purpose of CDN is to add a new layer of network architecture to the existing Internet, publish the content of the website to the "edge" of the network closest to the user, so that users can obtain the required content nearby, solve the Internet network congestion, and improve the response speed of users accessing websites. CDN is a combination of technologies, including several important parts: source station, cache server, and smart DNS.

Source station

The source station refers to the original site where the content is published. Adding, deleting and changing files on a website are all done on the source site; in addition, all objects captured by the cache server also come from the source site.

Cache Server

The cache server is a site resource directly provided to users for access. It consists of one or more servers. When a user initiates an access, his access request is located by the smart DNS to a cache server that is closer to him. If the content requested by the user happens to be in the cache, the content is directly returned to the user; if the content required for access is not cached, the cache server fetches the content from a neighboring cache server or directly from the source site, and then returns it to the user.

Smart DNS

Smart DNS is the core of the entire CDN technology. It mainly points its access request to a cache server that is closer to the user based on the source of the user, such as pointing the user request of Guangzhou Telecom to the cache server in the IDC computer room of Guangzhou Telecom. Through smart DNS resolution, users can access servers under the same service provider, eliminating the problem of slow access between the north and south networks in China, and achieve acceleration. The emergence of smart DNS has overturned the traditional practice of one domain name corresponding to one mirror, allowing users to access websites more conveniently.

As the business expands further, we need to split the application into business. Each business application is responsible for relatively independent business operations (so multiple services need to be opened to divide different businesses). Businesses communicate through messages or share databases. Therefore, a B business server is connected to the load server to distribute its business (such as users, orders, payments, etc.), and then the B-level business server communicates with the A business service area in real time through the message queue server to share data. There are many such A and B servers, and they can share data with each other.