Abstract
With the rapid increase of user access, load balancing in cloud data center has become an important factor affecting cluster stability. From the point of view of green scheduling, this paper proposed a virtual machine intelligent scheduling strategy based on machine learning algorithm to achieve load balancing of cloud data center. Firstly, a load forecasting algorithm based on genetic algorithm (SVR_GA), k-means clustering algorithm based on optimized min-max, and adaptive differential evolution algorithm (ESA_DE) to enhance local search ability are proposed to solve the load imbalance problem in cloud data center. The experimental results showed that compared with other classical algorithms, the proposed virtual machine scheduling strategy reduces the number of virtual machine migration by 94.5% and the energy consumption of cloud data center by 49.13%.
Highlights
Load balancing plays an important role in resource management of cloud data center
Most load balancing strategies are optimized from the following aspects: load balancing strategy based on server CPU and memory resource utilization, load balancing migration strategy based on service-level agreement (SLA), load balancing strategy based on network traffic prediction, load balancing strategy based on quality of service (QoS), load balancing strategy based on service response time prediction, and load balancing strategy based on cloud storage
In order to accurately predict the load state of servers, this paper studied the traditional algorithm of support vector regression, found out three parameters to solve the accuracy of the traditional support vector regression algorithm, and proposed a load prediction algorithm based on genetic algorithm optimization support vector regression (SVR_GA)
Summary
Load balancing plays an important role in resource management of cloud data center. It can improve the resource utilization of servers and prevent servers from being overloaded, and effectively reduce the migration frequency of virtual machines and avoid unnecessary waste of resources. Aiming at the problem of the cluster number and the selection of initial cluster centers in k-means algorithm, this paper proposed a k-means clustering algorithm based on optimized min-max This algorithm found virtual machines with less migration cost, network traffic, and performance interference from overloaded servers. The virtual machine found by clustering algorithm was migrated to the target server with the lowest migration cost, network traffic, and performance disturbance, so as to achieve load balancing of cloud data center. Dri represented the length of network links when virtual machine i migrated, and the value is related to the network topology of cloud data center, the location of the source server, and the target server in the network topology It could be calculated by formula (5):. The specific process is shown in Algorithm 4: 2. Development environment: Eclipse 4.5.1,JDK1.8.0_111
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have