Shulou(Shulou.com)06/02 Report--

This article focuses on "the specific use of strings in Redis", interested friends may wish to take a look. The method introduced in this paper is simple, fast and practical. Now let the editor take you to learn "the specific use of strings in Redis"!

Working with scen

one。 Caching function

When we do development, we all know that all the data in the project is obtained from the storage layer, that is, in the database. However, if all requests are obtained from the database, it will cause great pressure on the system, because obtaining data directly from the database will involve read and write operations in the database, and read and write operations in the database will consume system resources. So for most companies, a cache layer will be added to the architecture of the system, and most of the request data will be obtained from the cache layer first. if no data is found in the cache layer, it will be obtained from the storage layer, that is, in the database. The data obtained by the storage layer is then synchronized to the cache layer. As a result, most requests are found in the cache layer, which greatly reduces the pressure on the storage layer. The cache layer Redis is one of the solutions. Let's briefly simulate the process of a user requesting data.

two。 Count

Counting function is common in many websites, the most typical is the number of pages played. For a video website, due to the large number of visitors, if the database is directly modified every time someone visits, then such a large amount of concurrency and frequent database modifications will certainly put great pressure on the database. In extreme cases, it may lead to database downtime, etc., and because of the large number of visits, we should also consider the compatibility of multi-threads when developing. Otherwise, the data will be inaccurate. In addition, due to the large number of visits, it will also cause each user to request a longer return time, users may be slow to visit the site, and so on. This is not the case with Redis. First of all, Redis saves the data in memory, and the performance is significantly improved compared with the disk IO operation of the database. Secondly, Redis is a single-threaded architecture, so we don't have to do special multithreading for large concurrency. The third is that Redis provides a lot of commands that support atomic operations, which we can use directly without considering the details. Therefore, it is more appropriate to use Redis to achieve the counting function of the website or other business. It is important to note, however, that we save the counted data in Redis in order to perform database modifications infrequently. The final result of the data is still saved in the database (although Redis has persistence capabilities). So in the actual development, we can choose a certain point in time, when synchronizing the counting data in Redis to the database, most of them will use the way of timing scheduling to synchronize the data, of course, we can also consider other counting implementation.

three。 Share Session

We know that the user's login information is saved in the Session during the project development. when a user accesses the system, we first judge whether the user's Session has the user's information, and of course, whether it exceeds the Session expiration time. If there is, it is considered that the user has successfully logged in, so access is allowed, otherwise, the user is prompted to log in, or jump directly to the login. The above scenario is not a problem in a single architecture, but it is a problem in a distributed architecture. We know that the Session is stored on the server, that is, the Session is stored on the machine on which our server is deployed. The distributed way is to deploy the server to multiple machines. This will lead to a problem. Although the user has logged in successfully, due to load balancing and other reasons, the server that provides the service to the user and the server that logs in to the user are not on the same machine, so it will appear. Although the user has logged in successfully, we will still prompt the user that he or she has not logged in. This occurs because other machines do not have user Session information except that the user logs in to that machine. This is what the following figure shows:

Since the above scenario has a problem in the distribution, I have to find a way to solve it. There are many ways to solve it. Here, only one solution is introduced, that is, using Redis solution. When the user logs in successfully, instead of saving the Session information to the local server, we save it to Redis. In this way, no matter which server logs in first, there is only one copy of the user's Session information, that is, the one saved in Redis. In this way, when other servers determine whether the user is logged in, they all obtain Session information from the Redis. If there is a user's Session information in the Redis, the user must have logged in successfully. Otherwise, the user has not logged in, or the login fails. This solves the problem of distributed user login. This is shown in the following figure:

These are all usage scenarios for string types in Redis, but they are much more than that in actual development. As long as we are proficient in using related commands of string types in Redis, we can solve many complex problems in our development.

At this point, I believe you have a deeper understanding of "the specific use of strings in Redis". You might as well do it in practice. Here is the website, more related content can enter the relevant channels to inquire, follow us, continue to learn!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.