The Study On Resource Scheduling Mechanism Related To Energy Efficiency Of Cloud Computing Platform

Recently,the demand of computing resources becomes more and more urgent.The computing and storage capacity of general computers constrain the applications of modern official business.The appearance of cloud computing provides a new solution for the requirement of massive computing.However,the cloud computing data center not only consumes a lot of electricity,but also produces a lot of carbon dioxide,which leads to the resource waste and the aggravation of the greenhouse effect.Herein,it is very important to reduce the energy consumption of data center.This paper mainly studies the energy consumption problem in cloud computing platform,through the design of appropriate resource scheduling algorithm to improve energy efficiency.The main work is as follows.In order to reduce the energy efficiency of data centers due to the large number of low-load hosts in the computing process of cloud computing platforms,an energy efficiency-related ant colony algorithm(ACO)is proposed to reduce the number of low-load hosts and improve the energy efficiency of cloud computing centers.Firstly,the energy related cloud computing resource scheduling problem is modelled,and the scheduling problem is mapped to the ACO problem.Then,the migration probability and pheromone concentration update formula are defined,and different schemes are adopted to deal with the idle hosts and active hosts,which solves the problem of adaptability of the existing ant colony algorithms under specific conditions.In addition,the traditional Best Fit Decreasing(BFD)algorithm is enhanced for the initialization to solve the problem of slow convergence of the ant colony algorithms.Finally,simulation experiments on CloudSim show that the energy consumption in two scenarios is reduced about 11.13% and 90.20% compared with the other solutions.To combat the problem that a small number of virtual machines occupy idle hosts in cloud computing platform,an energy efficiency-related time-aware ant colony algorithm is proposed,which reduces the overall energy consumption of cloud computing platform by combining the task waiting delay property.Firstly,a delay allocation scheme is proposed for the time-aware virtual machines.Then the service availability about the service level agreement(SLA)is introduced as an indicator for the constrained supervision.After that,two key parameters are defined to determine the set of virtual machines suitable for delay allocation strategy.Finally,lots of simulations have been carried out on the CloudSim platform to validate the proposed solutions.The validity of the results,availability of SLA services and stability of the algorithm are taken as indicators to determine the range of the key parameters.The results show that this method further reduces the total energy consumption about 7.31%.