eduzhai > Applied Sciences > Engineering >

Nature Inspired Optimization Techniques for Cloud Scheduling Problem

  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: With the affluence of large number of users and providing necessary services to them and to meet their CSAT (Customer Satisfaction) or services is still challenging task. Day by day the user’s data size in cloud (network of networks) is increasing enormously. With the help of huge amount of systems in the world it becomes a primary goal to scheduling a job. But they are still expected to provide new innovative IT solutions to meet business requirements & ensure that the IT infrastructures is agile & scalable enough to meet is to embrace cloud computing. It is not something you can avoid anymore, because a lot of companies across diverse industries are migrating to the cloud model & benefitting from it. For a larger enterprise with significant investment in their on premise IT infrastructure a cloud with good scheduling algorithm make more sense, where part of their infrastructures is in the public cloud or private cloud and part on premise.In this paper various optimization techniques such as Genetic algorithm, Multi queue, Ant Colony optimization, particle swarm optimization, selective breeding, taboo search, Lion optimization techniques, firefly algorithms for cloud scheduling problems are discussed and simulated few algorithms like ACO, PSO, GA, LOA and compared its result with firefly algorithm in order to improve the performance of the cloud scheduling and workflow management in cloud. Users are allowed to submit their jobs & schedulers are allowed to allocate the resources, schedule the jobs based on certain job constraints. the results are obtained, the firefly (FA) in most cases it provides excellent results in fast convergence and global optima success and also Firefly is compared with all other cases of various meta heuristics.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...