Scheduling task publishing system and method

Abstract

The invention discloses a scheduling task publishing system and method. The scheduling task publishing method comprises the following steps: S1, delivering a scheduling task to be published to a corresponding task executor to be responsible for scheduling execution; s2, configuring hardware scheduling resources required by operation for the scheduling task; s3, associating the scheduling tasks according to a logic relationship, and combining the scheduling tasks into scheduling jobs; s4, establishing a scheduling strategy for each scheduling task in the scheduling job; and S5, publishing the scheduling job and starting execution. According to the publishing method, the design of the task executor is adopted; multiple task types and calling modes can be supported, internal implementation ofprograms of different languages and different types is shielded, only how to execute and call the tasks needs to be paid attention to, and even after program codes are changed and set and other languages are adopted for reconstruction, the scheduling tasks can still be reissued only by simply adjusting calling configuration information.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a dispatching system and method for dispatching tasks. Background technique [0002] With the continuous development of social informatization and software technology, the stacked software architecture of traditional independent systems can no longer meet the ever-changing business needs and the increasingly large volume structure and high maintenance costs, which often affect the whole body because of the slightest influence. Therefore, more and more software has shifted from monolithic applications to microservice architecture and big data platform technology framework. While this transformation brings advantages and convenience, problems also arise spontaneously. Due to the variety of development languages, the backgrounds of developers are very different, which produces many different types of programs (tasks) running on the big data platform. Communication, reducing...

AI Technical Summary

Problems solved by technology

Program developers need to consider not only the implementation logic of scheduling tasks, but also the compatibility issues caused by scheduling tasks that may be deployed on different operating systems, as well as the specific scheduling strategies of tasks, etc., resulting in a significant development workload. Rise, the transformation time cycle is prolonged, the input resources and costs are increased, the efficiency is low, maintenance is difficult, and it is impossible to respond quickly to the demand

Images