Research On Collaborative Allocation Of Computing,storage And Transport Resources For Cloud-network Integration

Nowadays,cloud-network integration has become an important direction of network development and will bring about new industrial changes.In the context of cloud-network integration,cloud data centers can provide users with stable computing and storage resources.Optical networks which connect users with the cloud and different cloud data centers can provide reliable bandwidth transport resources.From the perspective of user needs and cloud-network integration,how to efficiently coordinate computing,storage,and transport resources to meet user needs and improve cloud-network resource utilization has become an urgent problem to be solved.This paper focuses on the three perspectives of "cloud network architecture,latency constraint,load balancing," and designs a network control architecture for cloud network integration.From the perspectives of latency constraint and load balancing,two corresponding algorithms for collaborative allocation of computing storage and transport resources are proposed.The main innovations of this paper are as follows:First,in view of the architecture design of cloud-network integration,a network control architecture for cloud-network integration is proposed.The architecture is based on software-defined network and network function virtualization technologies.The server orchestrates and deploys different cloud services,and combines with different resource allocation strategies to achieve efficient allocation of computing,storage and transport resources.Second,aiming at the heterogeneous resource allocation problem of latency-sensitive cloud services under the background of cloud-network integration,the end-to-end latency model of the services is given by analyzing the requirements of low-latency services and the end-to-end latency of cloud services.This paper proposes a collaborative allocation algorithm for computing storage and transport resources based on latency constraints.The simulation results show that the algorithm can reduce the service end-to-end delay by about 2ms compared with the benchmarkalgorithm,and can reduce the service blocking rate by 7%.It can improve the utilization rate of the three resources of computing,storage and transport.Thirdly,aiming at the load balancing problem of multi-cloud data centers and optical networks in the context of cloud-network integration,a collaborative allocation algorithm for computing,storage,and transport resources based on load balancing is proposed by introducing computing,storage,and transport resource availability expressions and resource load models.The simulation results show that the proposed algorithm can reduce the blocking rate by nearly 4%compared with the benchmark algorithm,and reduce the average hops by about 0.5 hops,and can improve the utilization of computing and storage resources.