Method, device and system for implementing task in cluster circumstance

Abstract

The invention relates to a method, a device and a system for realizing a task under a cluster environment. In the embodiment of the invention, the designated application in the cluster environment obtains the task from a database, and the obtained task is loaded; a task example is created, and is stored in a memory of a server where the designated application belongs to, thereby avoiding the circumstances that the task in the database is loaded repeatedly in the cluster environment.

Description

Technical field [0001] The present invention relates to the field of communication technology, and in particular to a method, system and device for realizing tasks in a cluster environment. Background technique [0002] In order to improve the disaster tolerance and performance of the system, cluster technology is increasingly used. However, once many functions of the previous stand-alone version are migrated to a cluster environment, problems may arise. Such as the realization of the task function, because the application of each server node in the cluster shares the database (as attached figure 1 As shown), therefore, multiple applications in different servers may obtain the same task in the same database at the same time (such as the startup or restart process), which will cause a task to be repeatedly loaded by multiple applications, resulting in one The task was mistakenly turned into multiple tasks in execution. [0003] In order to avoid the above error scenarios, the exi...

AI Technical Summary

Problems solved by technology

[0005] The structure of the problem-solving system adopted in the above-mentioned prior art is complex, and it is difficult to ensure that the startup state of the task can be accurately marked. There may be a delay between the two operations of task and marking, and another application may also get the same task during the delay period, which will cause a task to be launched by multiple applications

[0006] Further, in the above prior art, if a certain application server is abnormal or shut down, the existing system cannot automatically migrate tasks to other normal working application servers, so that tasks in the abnormal or shut down system cannot be application

[0007] In addition, in the above-mentioned prior art, when a certain application starts first, all tasks may be loaded on a certain application, causing the application to become the performance bottleneck of the entire cluster

Images