Study On Resourse Scheduling And Routing Problems In Deterministic Computing First Network

Computing first network is a new type of network structure that unifies the control and scheduling of the three structures of cloud,network,and edge to achieve efficient transmission,processing,and flexible routing of network resources.Deterministic transmission guarantees the controllability and accuracy of the time delay,jitter and other indicators during the transmission process of the business,and the realization of deterministic transmission in the computing first network will help the efficient and orderly transmission of data.Therefore,the realization of deterministic transmission in the computing first network is of great significance to improve the quality of user service.Designing a reasonable deterministic computing first network resource scheduling and routing scheme can effectively reduce transmission delay,processing delay and improve transmission reliability.Aiming at the resource scheduling and routing problems in the deterministic computing first network,this paper proposes the joint intelligent optimization scheme of deterministic scheduling and routing in the computing first network and the multi-path routing optimization scheme of heterogeneous computility in the computing first network,and Simulation experiments are carried out according to the designed system model and algorithm.The specific work of the paper is as follows:1)The computing first network integrates heterogeneous computility information with the network,which improves resource utilization and network transmission efficiency.The time-sensitive network ensures lowlatency and high-reliability transmission.The fusion of the two can achieve high-efficiency deterministic forwarding.The resource scheduling and routing planning in the integrated decision-making computing first network and the gating arrangement in the time-sensitive network will have problems such as too many decision variables,high computational complexity,and insufficient optimization performance.Aiming at the above problems,a fusion architecture based on IEEE 802.1 Qbv for gating arrangement,computing first network routing planning,and computility resource scheduling is proposed.Based on deep reinforcement learning,an improved RBDQN(reward back deep Q network)algorithm is proposed to optimize the gate control,and a greedy algorithm is used to assist routing path planning.The algorithm uses the average delay,energy consumption and user satisfaction as multiple optimization indicators to establish utility functions.The simulation results show that compared with the genetic algorithm,RBDQN can reduce the convergence time of small-scale scheduling problems by more than 1 times,and it can reduce the convergence time by dozens of times for multi-service and multi-node computing first network problems.At the same time,the algorithm can prevent the model from falling into local optimum.Compared with the traditional DQN(deep Q network),the decision result improves the performance of the utility function index by more than 10%,and the convergence time under the same index decreases by about 50%.2)Heterogeneous computility is the embodiment of business computing diversity in the era of heterogeneous computing.Reasonable analysis of heterogeneous computility in the computing first network can greatly improve business processing efficiency.At the same time,because user experience and service quality are closely related,ensuring computility Reliability in deterministic transmission in the network is also very important.In view of the above problems,this paper proposes a multipath routing optimization scheme for heterogeneous computility in a computing first network,and builds a heterogeneous computility measurement model and a multi-path routing planning model according to actual business needs.According to the nature of whether the user service can be split,combined with the K-routing-cooperation(K-R-C)mechanism and K-ACO algorithm proposed in this paper,the multi-path routing path planning is realized.Through simulation experiments,the single-path routing algorithm,multi-path random algorithm,KSP algorithm and K-A*algorithm are compared under different indicators.The results show that the reliability of the algorithm in this paper is the best under the condition that the node/link failure rate increases.In addition,the transmission delay expectation and the search convergence time are relatively minimum.In terms of processing delay indicators,it is verified that the K-R-C mechanism can reasonably arrange scheduling nodes with heterogeneous computility and effectively reduce processing delays.