Combination Auction-Based Online Scheduling for the Deterministic IP Networks
Deterministic Internet protocol networking uses frequency synchronization to improve the network scalability.However,frequency synchronization introduces a more complex forwarding model,accompanied by non-convex and non-differentiable operations,which increases the complexity of scheduling problems.Moreover,most scheduling schemes still rely on offline mechanisms or online greedy mechanisms,which not only make the algorithm complex,but also difficult to ensure the access of important time-sensitive services that arrive in the future.To address these issues,an online flow scheduling algorithm is proposed based on the combinatorial auction,where network resources are auction items,the flows are bidders,and the network is the auctioneer.Before all time-sensitive streams arrive,it is necessary to complete the offline resource enumeration operation with non-convex and non-differentiable operations.When the network is put into operation,it only needs to select resources from the pre-calculated feasible set to allocate to the newly arrived time-sensitive services,which reduces the computational complexity of the online part.Meanwhile,based on the framework of combinatorial auction,a linearly increasing resource pricing function is designed to ensure that key time-sensitive services can obtain network access opportunities.Finally,simulation experiments show that the proposed algorithm can provide efficient transmission services for important time-sensitive services in large-scale networks.
deterministic Internet protocol networkingonline flow schedulingauction theory
NETL
NSTL
万方数据
