Method and system for realizing delay queue through multiple strategies

Abstract

The invention discloses a method and system for realizing a delay queue through multiple strategies, and belongs to the technical field of delay queues, redis is integrated on the basis of a system, zset adding, deleting and querying operations of the redis are realized, and the method comprises the following steps: a message instance, one message comprising message topic, a message body, delay time, consumption times and consumption exception processing types; generating a message, packaging a message instance according to service requirements, setting top, a message body, delay time, consumption times and consumption exception processing types, and putting the message into zset through a redis tool method; message consumption; and the consumption strategies comprise a consumption method transverse multi-strategy and a consumption method longitudinal multi-strategy. According to the invention, the distributed and lightweight class can be realized, the disaster tolerance is relatively good, and the research and development cost and the maintenance cost can be reduced.

Description

technical field [0001] The invention relates to the technical field of delay queues, in particular to a method and system for implementing delay queues with multiple strategies. Background technique [0002] Delay queues are generally implemented by using scheduled tasks, DelayQueue, redis expiration callback, rabbitmq, etc. However, existing technologies either cannot implement distributed calls, or have poor disaster tolerance, or require heavyweight middleware, which is not easy to expand. [0003] The traditional delay queue is generally implemented by rabbitmq, but rabbitmq is relatively heavy. If the current business system uses rabbitmq, it can be considered, but if not, it is more cumbersome to use rabbitmq. Most of the DelayQueue and timing tasks that come with java do not support distributed , and non-persistent, the expiration callback of redis supports distributed but poor disaster recovery, which may easily lead to message loss. Contents of the invention [0...

AI Technical Summary

Problems solved by technology

[0002] To realize the delay queue, timing tasks, DelayQueue, redis expiration callback, rabbitmq, etc. are generally used, but the existing technology either cannot realize distributed calls, or has poor disaster tolerance, or needs to introduce relatively heavyweight middleware, which is not easy to expand

[0003] The traditional delay queue is generally implemented by rabbitmq, but rabbitmq is relatively heavy. If the current business system uses rabbitmq, it can be considered, but if not, it is more cumbersome to use rabbitmq. Most of the DelayQueue and timing tasks that come with java do not support distributed , and non-persistent, the expiration callback of redis supports distributed but poor disaster tolerance, which may easily lead to message loss

Images