Discrete Particle Swarm Scheduling Algorithm Based on Oscillating Inertial Weights

Abstract

The invention discloses a discrete particle swarm scheduling algorithm based on oscillating inertia weight, which includes the following steps: S1, initialize the parameters of each particle in the particle swarm; S2, calculate the completion time of each particle, and find out the overall optimal solution g best and the individual optimal solution p best ; S3, updating the speed and position of each particle: S4, judging whether the number of iterations has been reached, if the number of iterations is reached, then terminate and output the result; if the number of iterations is not reached, then loop into step S2. The invention greatly shortens the time of task scheduling in cloud computing through the discrete particle swarm scheduling algorithm of oscillating inertia weight, improves the feedback speed to user requests, and improves the service quality of the cloud platform.

Description

technical field [0001] The invention relates to the field of cloud computing resource scheduling, in particular to a discrete particle swarm scheduling algorithm based on an oscillating inertia weight. Background technique [0002] With the development of Internet technology and the explosive growth of mobile Internet users, the Internet is further integrated into all aspects of users' lives. This new situation also brings new challenges to network providers. Meeting users' requirements for network speed and security has become an urgent problem for network providers, and the continuous growth of network bandwidth has made it possible to access non-local computing services through the network. Cloud computing was born in this context. . It is a new computing model based on traditional computer technology and network technology, and integrates technologies such as virtualization technology, load balancing, distributed computing, parallel computing, and network storage. In t...

AI Technical Summary

Problems solved by technology

However, it is an NP-hard problem to map many subtasks to resource nodes. There are certain problems in the scheduling algorithm adopted by the cloud computing platform at present: such as single-queue scheduling, which is simple but has low resource utilization; fair scheduling, which supports job classification scheduling, but the node Load balancing is not considered; capacity scheduling supports parallel execution of multiple jobs, but queue setting and queue selection cannot be performed automatically

Images