How does redis implement message queue?

Change the request processing of the instant server to asynchronous processing to relieve the pressure on the server and achieve sequential data acquisition. This article mainly shares with you how redis implements message queues, hoping to help everyone.

Message queue, in layman’s terms, is a container that temporarily saves messages during the message transmission process. It can transmit data between applications on different platforms and different languages, and Data writing can be implemented asynchronously, which can play a good role in dealing with large concurrency problems.

Let me talk about the scenario where I use message queue now: my system needs to receive requests from clients in real time (periodic requests), and save the requested data in the database. The amount of data requested each time is very small. However, due to the need to frequently operate the database, the system encountered performance bottlenecks. The original idea was to store this data directly in redis, but if you think about it carefully, although the amount of data each time is not large, because the data is cyclical and will continue to grow, although redis also has a persistence function, if you rely on persistence If you save it to the hard disk, you will lose the advantage of redis as an in-memory database, so you finally decided to use a message queue.

The method is as follows:

The data requested each time is written directly to the message queue, and then a response is given to the client. A thread is opened in the background to monitor the message queue. Once data is written, the message queue is obtained. The data is written to the mysql database, so that real-time data can be written to the database asynchronously, avoiding the bottleneck of the system caused by direct operation of the database.

Then the question is, which message queue to use? The more mainstream message queues are now such as: RabbitMQ, ActiveMQ, etc., their technology is relatively mature and very efficient. However, considering that the project itself has used redis (for caching), and the amount of data each time is relatively small, redis is not only a key-value database, it supports rich data types, such as HashMap, Set, List, etc., among which The List can be used as a message queue, and the Redis List supports blocking commands such as blpop and brpop, which can fully meet my needs. The simple test code is as follows:

//产生数据 import redis.clients.jedis.Jedis; public class RedisProducer { /** * jedis操作List */ public static void main(String[] args){ Jedis jedis = new Jedis("192.168.10.209", 6379); for(int i = 0;i<10;i++) { jedis.lpush("informList","value_" + i); } jedis.close(); } } //消费数据 import java.util.List; import redis.clients.jedis.Jedis; public class RedisConsumer { /** * jedis操作List */ public static void main(String[] args){ ScheduleMQ mq = new ScheduleMQ(); mq.start(); } } class ScheduleMQ extends Thread { @Override public void run() { while(true) { Jedis jedis = new Jedis("192.168.10.209", 6379); //阻塞式brpop，List中无数据时阻塞 //参数0表示一直阻塞下去，直到List出现数据 List list = jedis.brpop(0, "informList"); for(String s : list) { System.out.println(s); } jedis.close(); } } }

Related recommendations:

PHP How to use redis message queue to publish Weibo

php implements message queue Class instance sharing

PHP Advanced Programming Message Queue

The above is the detailed content of How does redis implement message queue?. For more information, please follow other related articles on the PHP Chinese website!