DefinePK

Abstract

Cloud computing is a popular technology where all applications and files are hosted on a cloud. One of the most challenging issues in cloud computing is load balancing which needs to be investigated for its perfect realization. Resource Intensity Aware Load Balancing (RIAL) was proposed for load balancing in cloud computing. Based on the dynamic weight assignment to resources, the RIAL selected the Virtual Machines (VMs) from heavily loaded PMs to migrate out and placed those VMs in lightly loaded destination PMs. An Improved RIAL was proposed to consider both the lightly loaded and heavily loaded PMs for selection of destination PMs. However, some important measures such as power consumption, temperature, and traffic were not considered in IRIAL while the selection of migration VMs and destination PMs. So, Power Consumption Aware- Traffic Aware- IRIAL (PCA-TA-IRIAL) method was proposed which considered power consumption, temperature and traffic measures to select the migration VMs and destination PMs. For an optimal selection of migration VMs and destination PMs, optimization algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Artificial Plant Optimization (APO) are introduced in this paper. Based on the crossover and mutation process, GA optimally selects the migration VMs and the destination PMs. PSO algorithm optimally selects the migration VMs to the destination PMs by updating the position and velocity of each particle in the population based on the cost value. APO algorithm is inspired by a tree’s growing process. Based on the light intensity and photosynthesis, each branch of the tree in APO optimally selects the migration VMs and destination PMs. Thus the optimization algorithms optimally map the migration VMs and the destination PMs effectively for load balancing.