Distributed service primary key generation method based on Mysql

Abstract

The invention relates to a Mysql-based distributed service primary key generation method, which comprises the following steps of: firstly selecting the time of an updating period which is greater than the time of a current machine, then selecting a large average value, and finally caching the workId data of the current service to a local file. According to the Mysql-based distributed service primary key generation method, the influence on service performance is relatively small, multiple services and multiple scenes can be supported to acquire unique representation IDs at the same time, related isolation can be carried out based on service identifiers, ID generation rules among the services do not influence one another, and high availability and high performance of micro-service services are supported. The local service is independently generated and can be correspondingly decoupled with the Mysql, and even if the Mysql cluster breaks down, the local service can work independently and is not affected by the Mysql cluster.

Description

technical field [0001] The invention relates to the technical field of distributed systems, in particular to a method for generating a distributed business primary key based on Mysql. Background technique [0002] In traditional relational databases, such as MySQL, the database itself has its own self-incrementing primary key generation mechanism. However, in a distributed environment, the single-table auto-incrementing primary key cannot be used after the data is horizontally split due to sub-database and table division. Therefore, we need a A globally unique id generation strategy is used as a distributed primary key. [0003] In a distributed system, the business composition is basically composed of multiple microservices with different functions. Each microservice has its own business process, and often needs to uniquely identify a large amount of data and messages. As the data grows day by day, the data sub-database After the table is divided, a unique ID is required t...

AI Technical Summary

Problems solved by technology

[0008] In addition to the above requirements, the ID generation function should try to ensure QPS, low average delay and other characteristics, and try not to affect the performance and high availability of the main business process. However, the auto-increment ID of the database in the existing distributed system cannot meet the requirements. need

Images