Performability Analysis Of Fog Data Center

Fog computing is a concept that is attracting more and more attention.Today,as more and more companies realize its benefits,a variety of solutions are now available on the market.Fog Computing is a distributed decentralized Infrastructure and a mediator between hardware and remote servers.It regulates which information should be sent to the server and which can be processed locally.In this way,fog is an intelligent gateway that offloads clouds enabling more efficient data storage processing and analysis.Fog computing approach is used to deploy a large number of data centers,spread over several sites,and placed closest to the users.These data centers called fog sites have a small number of servers(about ten)and respond to requests.Many applications can benefit from this infrastructure,such as the data processing from sensors,in a city or smart factory context,to deploy virtual network devices closer to users,within cellular networks or event to enable the development of virtual reality solutions.The Infrastructure as a Service(IaaS),a kind of basic fog services is provided to customers in the form of virtual machines(VMs).The increasing demands for laaS fog services require the performability analysis of fog infrastructure,and for that,the analytic modeling is one of the effective approaches.This thesis aims to develop a monolithic model by using continuous time Markov chain(CTMC)for fog data centers,which(1)consists of active and standby physical devices(PDs),(2)allows physical device migration among active and standby physical device pools,(3)all jobs are homogeneous,and(4)a running job could continue to run by using idle active physical devices when the physical device working for this job fails.A monolithic CTMC model for performability analysis could be used to verify the scalable approximate model.In this thesis,I present the details of state transition rules of the proposed model and the formula for computing metrics,including mean job response time,job rejection probability and job immediate service probability.Numerical analysis and simulations are carried out to verify the accuracy of the proposed model.