Cloud robot resource scheduling method and system based on task classification and time sequence prediction

Abstract

The invention provides a cloud robot resource scheduling method and system based on task classification and time sequence prediction, and relates to the technical field of cloud computing resource scheduling, and the method comprises the steps: S1, obtaining a job task sequence of a cloud robot; S2, classifying the tasks according to the dependency relationship between the cloud robot operation tasks and the cloud virtual machine; S3, obtaining an estimated value of the job duration of each job task; S4, converting cloud robot operation task scheduling into a workflow scheduling problem expressed by a directed acyclic graph (DAG); S5, simplifying the DAG generated in the step S4 according to the characteristics of the job task; and S6, solving the final DAG by using various static scheduling methods. According to the method, static analysis can be finally carried out on the operation task of the cloud robot through some preprocessing means, and a better solution for the scheduling problem can be found from the global perspective.

Description

technical field [0001] The present invention relates to the technical field of cloud computing resource scheduling, in particular to a cloud robot resource scheduling method and system based on task classification and timing prediction. Background technique [0002] The resource scheduling problem of the cloud robot aimed at by the present invention mainly considers the mapping problem between cloud job tasks and different virtual machines in the cloud. [0003] The scheduling problem of cloud computing tasks is a hot topic in the direction of cloud computing. In cloud computing tasks, by reasonably mapping a large number of cloud job tasks to a limited number of virtual machines, the earliest completion of the workflow or workload balancing can be achieved. and other service quality indicators. However, there is little research on resource scheduling methods specifically for cloud robots. The biggest difference between the resource scheduling method of cloud robots and th...

AI Technical Summary

Problems solved by technology

[0007] Although this kind of dynamic scheduling scheme can also solve the resource scheduling problem of cloud robots to a certain extent, compared with static methods, dynamic methods have less information, so they can only consider the optimal solutions of each part, and cannot learn from the overall situation. Looking for the best solution from a different perspective

Images