Distributed call system for timed tasks

Abstract

The present invention provides a distributed call system for timed tasks. The system includes: a client cluster, including a plurality of client nodes, which are used to obtain task registration information and send it to the server node, and also send heartbeat information to integrate the user's operation The command is sent to the server node; the server cluster, including multiple server nodes, is used to create a scheduled task; when the scheduled task trigger time is reached, the task execution node is determined, and the task execution node executes the scheduled task; a task execution monitoring interface is provided to receive customer End node operation instructions and perform operations on scheduled tasks; ZooKeeper clusters are used to save the status information of client nodes and server nodes; MongoDB clusters are used to save task registration information and save task execution nodes and servers corresponding to scheduled tasks The ID of the node. The invention avoids a single point of failure, saves local service resources, and realizes the supervision of the task execution process.

Description

technical field [0001] The invention relates to the technical field of computers, in particular to a distributed call system for timed tasks. Background technique [0002] In the prior art, when configuring a scheduled task, the scheduled task is generally configured through the Spring annotation @Scheduled, for example: the configured scheduled task is @Scheduled(cron="0 0 0 / 1**?"), that is, every hour Execute a scheduled task once, and the configured scheduled task is executed locally on the server. [0003] Because the task execution process is invisible, it is impossible to control the pre-perception of the execution progress of the task and whether the execution is terminated due to abnormality; and because the task is executed locally, that is, the execution of the task is in the same JRE (JavaRuntime Environment, Java Runtime Environment) as the application. ), the impact on the performance of the application itself cannot be controlled, and there is a high probabili...

AI Technical Summary

Problems solved by technology

[0003] Since the task execution process is invisible, it is impossible to control the pre-perception of the execution progress of the task and whether the execution is terminated due to abnormality; and because the task is executed locally, that is, the execution of the task is in the same JRE (JavaRuntime Environment, Java Runtime Environment) as the application. ), the impact on the performance of the application itself cannot be controlled, and there is a high probability that the performance of the application service will drop sharply due to abnormal timing tasks; during the execution of the scheduled tasks, manual intervention cannot be performed for batch data processing; and due to the Created locally and executed jointly on all local nodes, resulting in inability to fully utilize cluster resources

[0004] To sum up, the scheduled tasks configured through the Spring annotation @Scheduled method have the following defects: prone to single point of failure; consume local service resources; difficult to supervise the task execution process; cannot make full use of cluster resources

Images