BUDGET-BASED SERVICE ALLOCATION AND LOAD BALANCING WITH MEMORY DE-DUPLICATION FOR AUTO SCALING IN CLOUD COMPUTING
Main Article Content
Automatic Scaling provides the opportunity for the applications where their resource utilization can be automatically scaled up and scaled down by the cloud service provider. But achievement of high demand satisfaction ratio is a difficult problem. So, the existing research Class-Constrained Bin Packing (CCBP) problem is formulated in which every server is denoted as bin and every class denoted as an application. An efficient Semi-Online Color Set algorithm (SOCS) is presented for solving the CCBP problem. But the drawback in this method is the demand satisfaction is less when the resources become tight and load balancing among the VMs is not considered that is essential for improving efficiency. In this manuscript an innovative technique called Optimal Selection of Cloud Service with Price for QoS Concession Budget-based Service Allocation and Load Balancing with Memory De-Duplication (OSCSPQC-BSA-LBMD) method is introduced for improving demand satisfaction ratio and achieves load balancing. In this method, the customers are sorted according to their budgets and allocate the resources for the high premium customers. If the load of a particular server is increased, live migration is applied for balancing load among the servers. For that, the statistical information like CPU utilization and memory consumption are gathered and assign the VMs to servers. During the live migration, in order to avoid unnecessary migrating memory pages the Memory De-Duplication method is used. In this method, the duplicate memory pages are identified and stored only once. Experimental results show that the proposed method achieves high demand satisfaction ratio and reduce the complexity.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.