Research On Resource Scheduling Technology In NFV Environment

In the context of the information age,emerging technologies are constantly evolving.As an infrastructure,the network carries more and more applications and services,which brings great impact and challenges to the traditional network.As a result,many closed network middleware devices are introduced into the network to improve performance,but this makes the network more complex and unable to meet the growing demands of applications and services.The emergence of network function virtualization(NFV)is an emerging network architecture that provides a new solution to the above problems,and deploys network functions in the form of software to replace traditional network middleware devices.NFV increases the flexibility of the network by decoupling network functions from hardware physical devices.In an NFV architecture,virtual network functions(VNFs)are typically deployed and serviced as a chain of service functions.Under the increasingly rich application requirements,how to deploy VNF and service function chains in the NFV environment and schedule resources has become an urgent problem to be solved.As an important indicator of quality of service(Quality of Service,QoS),the availability and latency of the service function chain are widely concerned in the resource scheduling problem of NFV environment.In recent years,the problem of resource scheduling in NFV environments has been extensively studied and many important results have been achieved.However,there are still some problems that need to be solved urgently in terms of improving the availability of the service function chain and ensuring latency in complex application scenarios.In view of the above two aspects,the main innovation points of this paper are as follows:(1)This paper proposes a resource optimization algorithm based on availability improvement.The algorithm consists of two backup strategies and an algorithm based on dynamic programming.The algorithm first defines the availability model,taking into account both VNF failures and physical equipment failures.On this basis,the algorithm proposes two backup strategies.One is a low availability priority strategy,which prioritizes providing backups for services with lower availability;the other is a backup service placement strategy,in which backup services are prioritized on the physical nodes of a service that is working.In addition,an algorithm.deployment service function chain based on dynamic programming is proposed,which effectively reduces resource consumption on the basis of improving availability.(2)This paper proposes a resource optimization algorithm based on delay guarantee.The algorithm consists of two parts:the service function chain request mapping algorithm and the virtual network function request adjustment algorithm.The service function chain request mapping algorithm maps the service function chain request to the physical node and satisfies the latency constraint;the virtual network function request adjustment algorithm appropriately adjusts the virtual network function request type on the physical node and merges the virtual network function request,thereby effectively reducing the basic resource overhead.