In the energy consumption of data centers,network devices account for the largest proportion of energy consumption.In order to reduce the energy consumption of network devices and improve the quality of network services.Based on the characteristics of different types of businesses,research has proposed virtual machine placement and routing scheduling algorithms based on improved genetic algorithms and business awareness to cater to different types of businesses.After testing,improved genetic algorithms can re-duce energy consumption by about 70％compared to polygon scanning conversion algorithms and guided filtering algorithms;Reduce packet latency and loss rates by approximately 22％and 64％;Increase network throughput by about 40％.Compared with greedy ran-dom adaptive Sears program and simulated annealing algorithm,the business aware virtual machine placement and routing scheduling algorithm saves about 20％and 18％of energy consumption in the access layer and aggregation layer switches,respectively,and re-duces packet latency by about 8％.It can be seen that both improved genetic algorithms and business aware virtual machine placement and routing scheduling algorithms can achieve energy-saving of network devices in corresponding types of services.
维普
万方数据