Distributed task scheduling method and system based on messaging middleware

Abstract

The invention discloses a distributed task scheduling method and a system based on messaging middleware, belonging to the field of task scheduling of a distributed system. The scheduling of distributed tasks is carried out via message queuing, the tasks to be implemented are stored in a task queue of a master node at first and are then uniformly distributed into available sub-nodes under the master node through load equalization, and independent task processes are started by the sub-nodes to implement the tasks. During scheduling and implementation of the tasks, the real-time monitoring of the states of the master node and the sub-nodes can be realized through node monitoring. In case of faults, the tasks to be implemented or the tasks under implementation can be switched to the spare master node or the available sub-nodes. By adopting the method and the system, the independence of the tasks can be ensured, and the efficiency of task implementation can be improved. Due to the fault switching mechanism, the reliability of the scheduling and implementation of the tasks can be further ensured.

Description

technical field [0001] The invention relates to the field of task scheduling of distributed systems, in particular to a distributed task scheduling method and system based on message middleware. Background technique [0002] With the increasing scope of computer applications, the application of distributed real-time systems is becoming more and more extensive. Among them, the problem of task scheduling is a key issue in distributed real-time systems. A reasonable task scheduling algorithm for a distributed real-time system is of great significance to exerting the parallel performance of the system, ensuring the schedulability of real-time tasks, and maintaining the load balance of the network. [0003] In the existing task scheduling framework, Quartz scheduling provides a simple and powerful task scheduling mechanism. Quartz allows developers to flexibly define the scheduling time of triggers, and can map triggers and tasks associatively, such as according to the time inte...

AI Technical Summary

Problems solved by technology

[0004] 1) Single-machine version single-process application, once downtime occurs, it is impossible to ensure that the task is triggered correctly;

[0005] 2) Since it is a single-process application, there will be a risk of memory overflow when the number of tasks is too high, and the execution efficiency is low;

[0007] The existence of the above-mentioned shortcomings makes the single linear task scheduling system more and more unable to meet the existing business needs. The present invention is aimed at the above-mentioned problems. On the basis of including all the characteristics of Quartz, it provides a message-based The distributed task scheduling method of software ensures the reliability of task scheduling and execution, and provides a failover mechanism to improve the efficiency of task execution

Images