Hybrid Load Balancing Algorithm In Cloud Computing

Gardie, Birhanu

dc.contributor.author	Gardie, Birhanu
dc.date.accessioned	2020-06-04T06:42:13Z
dc.date.available	2020-06-04T06:42:13Z
dc.date.issued	2020-02
dc.identifier.uri	http://hdl.handle.net/123456789/10877
dc.description.abstract	Cloud computing is a virtual pool of common computing resources, which has presented to the customer through the internet. It gives unlimited pay per use of computing resources virtually without burdening the user by managing the underlying computing infrastructure. Providing cloud computing service is not straight forward task it needs appropriately balancing of load on virtual machines to achieve optimum allocation of bandwidth, memory utilization, processing speed and instruction size between virtual machines in the data center. The cloud system resource should coordinate to provide users request response that needs intercommunication among different parts of the system, this leads to challenge in an imbalanced charge in the diversified networking system where some node get involved in over charge, some get in light charge and others might involve idle. To alleviate such problem service providers required to apply a solution on load balancing for allocation of tasks over datacenters, network, hard drives, physical hosts, and virtual machines across these virtual cloud centered resources. Applying genetic algorithm and fuzzy set theory separately have should not improve on iteration time. The aim of this study was load balancing in cloud computing using hybrid algorithm to improve the overall performance of cloud-based system. On this study hybrid of genetic algorithm and fuzzy set theory has implemented to get optimal load balance among virtual machines in the datacenter. This study has been simulated on ten datacenters, fifty virtual machine and timeshared virtual machine provisioning policy using CloudSim simulation toolkit. The experimental simulation result showed an average of execution time 2.1 and 1.5 milliseconds for genetic algorithm and hybrid genetic algorithm respectively. Resource utilization is found 90.1 % and 53.2 % for hybrid algorithm and genetic algorithm respectively. The hybrid algorithm found less imbalance value of jobs in VMs as compared with genetic algorithm. In conclusion, proposed hybrid algorithm has found highest resource utilization and lower execution time.	en_US
dc.language.iso	en	en_US
dc.subject	Information Technology	en_US
dc.title	Hybrid Load Balancing Algorithm In Cloud Computing	en_US
dc.type	Thesis	en_US