How to use Redis for MySQL caching

Redis caching is actually to put frequently accessed data into Redis, users query first to Redis when querying, execute SQL statement query if they do not find it, and synchronize the data to Redis at the same time. Redis only performs read operations and has a fast query speed in memory.

Using Redis for caching must solve two problems: first, determine what data structure to use to store data from MySQL; Once the data structure is determined, it is necessary to determine what identity to use as the key of the data.

MySQL stores data in tables that are made up of rows. Each time a select query is executed, MySQL returns a result set consisting of several rows. Redis has five data structures:

List, hash, string, set, sorted set, compared with several data structures, string and hash are more suitable data structures for storing rows, you can convert the data into json strings and store them in Redis.

Full traversal keys: keys pattern keys *

Some people say that the key equivalent of a library of relational data select * is almost disabled in production.

Regardless of whether the above is true or not, keys are definitely risky. Then let's change to a type of sleepy bright plan, when storing data. Save the key of the data, and also save it in redis to select the hash type, so you can get all the values directly through this hash when you take it, and it is very easy to use self-feeling!

This is a pitted method, which can be used in general situations where there is not much traffic, for example, the background system. However, in the scenario of large front-end user traffic, once the hit rate of the hot data cache fails, the request to the database will kill the system. Therefore, this strategy should not be employed.

1. Redis is an in-memory data storage service, so its speed is faster than MySQL.

2. Redis only supports basic data types such as string, hashmap, set, sortedset, etc., but does not support joint query, so it is suitable for caching.

3. Sometimes the amount of cached data is very large, if the service is down at this time, and the persistence function of Redis is enabled, and the service is restarted, the data will basically not be lost.

4. Redis can do memory sharing because it can be connected by multiple different clients.

5. As a cache for MySQL and other databases, it is to store part of the hot data in Redis first, or load it into Redis when you use it for the first time, and take it directly from Redis when you use it next time.

6. The data in Redis can be set to expire, if the data is not extended within a certain period of time, the data will be cleared from Redis after a certain time.

Redis is used to read and write data, and the queue processor is used to write data to MySQL on a regular basis.

At the same time, it is necessary to avoid conflicts, read all table key values from MySQL when Redis is started, and store them in Redis, and when writing data to Redis, automatically increment and read the Redis primary key, if MySQL fails to update, you need to clear the cache and synchronize the Redis primary key in time.