Research On Migration In User Mobile-Aware Service Function Chain

With network function virtualization expanding from network center to edge,the service function chain(SFC)will gradually approach users to provide lower delay and higher-quality services.User mobility reduces the quality of service provided by the mobile-aware SFC.Therefore,we must migrate the SFC to provide continuous services.Mobile networks where users’movement path are certain and the time users arriving at the destination nodes are estimable are common in daily life,however there are no papers on migration research in the user estimated movement scenario.Migration path and bandwidth are the influencing factors of migration time and occupied resources.Therefore,the thesis researches the routing and bandwidth allocation problems for mobile-aware SFC in the user estimable movement scenario.Firstly,for the user estimable movement scenario,we establish the estimation model of user arrival time to obtain the estimated arrival time.Utilizing users’ known moving path and estimated arrival time,the thesis proposes a softer migration strategy migrating the mobile-aware SFC in advance,to reduce the user waiting time for the migration completion.Then,in the research of routing and bandwidth allocation when a single SFC migrates independently,a path load adaptive routing and bandwidth allocation(PLARBA)algorithm is proposed to reduce migration failure rate.The algorithm selects the shortest path satisfying the constraint as migration path and measures the path load of migration path to dynamically allocate migration bandwidth.Experimental results show that compared with the soft migration strategy the softer migration strategy can reduce the user waiting time by more than 90%,compared with the two routing and bandwidth allocation algorithms the proposed PLARBA algorithm can reduce the migration failure rate by up to 75.4%.Finally,in the research of routing and bandwidth allocation when multiple SFCs migrate collaboratively,aiming at the problem that reducing the migration failure rate would increase the gap between the expected migration completion time and the user’s actual arrival time,this thesis puts forward the routing and bandwidth allocation based on simulated annealing(SARBA)algorithm,to balance the overall failure rate of migration and time difference between migration completion and user arrival.Combined with simulated annealing algorithm with global convergence,SARBA algorithm takes the migration results of the PLARBA algorithm as the initial solution.As the temperature dropping,SARBA algorithm jumps out of local optimal solution to obtain the overall optimal migration route and bandwidth allocation scheme.Simulation results illustrate that compared with PLARBA algorithm,the SARBA algorithm can further reduce the overall migration failure rate by 75.2%at most,reduce the time difference between migration completion and user arrival by at most 28.0%,and decrease the bandwidth resource waste by nearly half.