关键词: 数据中心网络,失效恢复,网络结构,可用性

Abstract: In recent years,with the rapid development of cloud computing technology,the data center network which acts as the underlying infrastructure of cloud computing,plays an increasingly important role．But there are some failures in the network,which requires a strong and effective recovery mechanism to ensure data center available．The network structure plays an important role in the availability of the data center network．Data center network can provide various cloud computing services．As a result,data center networking has recently been a hot research topic in both academia and industry．A fundamental challenge in this research is the design of the data center network that interconnects the massive number of servers,and provides efficient and fault-tolerant routing service to upper-layer applications．In response to this challenge,the research community has begun exploring novel interconnect topologies including Fat-Tree,DCell,and BCube,etc．Extention of the proposed solutions is either too fast (i.e.,double exponentially) or too slow,they suffer from performance bottlenecks,or can be quite costly in both routing and construction．This paper proposed a cost-effective,gracefully scalable,and fault-tolerant data center interconnect termed DCNS that combines the advantages of both DCell and BCube architectures while avoiding their limitations．We then proposed fault-tolerant and routing mechanisms for DCNS．Finally,we proposed a comprehensive benchmarking environment that can be used for accurately and practically evaluating and testing the proposed data center architecture．Experimental results show that DCNS can better meet requirements of the data center network architecture and ensure system availability.

Key words: Data center network,Failure recovery,Network structure,Availability