I have been learning about caching in the last two days. Everyone knows that caching is to solve the problem of high concurrency and massive data. Mysql mainly becomes a performance bottleneck. The traditional method is to check the cache first, and then check the DB. Because the cache is used, the overall design of the program There will be no complicated table sql, etc. (this article is mainly for Internet C-side projects, b-side projects are mainly logically complex, and the concurrency is not high), and db is also stored in the form of key and value, mostly in the form of primary key query, aggregation, filtering, etc. The operation is all carried out in the code, because mysql cannot be linearly expanded, and the java application can be expanded horizontally, of course, it cannot be infinite, otherwise the number of connections to mysql cannot be supported.
During the viewing process, I found that Alibaba Cloud PolarDB is fully compatible with mysql and can be flexibly expanded. The performance is several times that of mysql, and it can handle complex SQL and so on. . . .
For a moment, I feel that in the future, programmers may not even need SQL optimization. No matter how bad the SQL is, they can return quickly. In the future, programmers will only engage in business CRUD full-time, sad.