{"title":"通过本地重路由放置高可用的虚拟网络功能","authors":"Xiaojun Shang, Zhenhua Li, Yuanyuan Yang","doi":"10.1109/MASS.2018.00022","DOIUrl":null,"url":null,"abstract":"The recent development of network function virtualization decouples network functions from dedicated hardware. Thus, virtual network functions (VNFs) can be distributed onto shared and virtualized platforms held by multiple data centers in various network locations. The architecture of the distributed VNFs interconnected by virtual links is denoted as the Service Function Chain (SFC). SFC has the potential to significantly reduce the opening and operating cost of the network services while improving the flexibility. However, the availability of SFC on chained up data centers is always inferior to that of network functions running on a single data center because the failure in any data center on the chain may affect its availability. To improve the availabilities of SFCs, in this paper we propose a local rerouting strategy to bypass the failed data centers on SFCs using locally rerouted paths (LRPs). Furthermore, we formulate an optimization model to minimize the maximum load on links while deploying SFCs and LRPs into the network. Thus, we reduce the potential risk of congested links caused by the local rerouting strategy. We then propose a randomized rounding approximation algorithm to solve the optimization problem, preserving the competitive ratio of O(logn) for the model. We also propose a fast heuristic algorithm to improve the efficiency. Our extensive simulation results show that the proposed algorithms can provide highly available SFCs with more balanced link load.","PeriodicalId":146214,"journal":{"name":"2018 IEEE 15th International Conference on Mobile Ad Hoc and Sensor Systems (MASS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":"{\"title\":\"Placement of Highly Available Virtual Network Functions Through Local Rerouting\",\"authors\":\"Xiaojun Shang, Zhenhua Li, Yuanyuan Yang\",\"doi\":\"10.1109/MASS.2018.00022\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The recent development of network function virtualization decouples network functions from dedicated hardware. Thus, virtual network functions (VNFs) can be distributed onto shared and virtualized platforms held by multiple data centers in various network locations. The architecture of the distributed VNFs interconnected by virtual links is denoted as the Service Function Chain (SFC). SFC has the potential to significantly reduce the opening and operating cost of the network services while improving the flexibility. However, the availability of SFC on chained up data centers is always inferior to that of network functions running on a single data center because the failure in any data center on the chain may affect its availability. To improve the availabilities of SFCs, in this paper we propose a local rerouting strategy to bypass the failed data centers on SFCs using locally rerouted paths (LRPs). Furthermore, we formulate an optimization model to minimize the maximum load on links while deploying SFCs and LRPs into the network. Thus, we reduce the potential risk of congested links caused by the local rerouting strategy. We then propose a randomized rounding approximation algorithm to solve the optimization problem, preserving the competitive ratio of O(logn) for the model. We also propose a fast heuristic algorithm to improve the efficiency. Our extensive simulation results show that the proposed algorithms can provide highly available SFCs with more balanced link load.\",\"PeriodicalId\":146214,\"journal\":{\"name\":\"2018 IEEE 15th International Conference on Mobile Ad Hoc and Sensor Systems (MASS)\",\"volume\":\"6 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"9\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 IEEE 15th International Conference on Mobile Ad Hoc and Sensor Systems (MASS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/MASS.2018.00022\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE 15th International Conference on Mobile Ad Hoc and Sensor Systems (MASS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MASS.2018.00022","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 9
摘要
网络功能虚拟化的最新发展将网络功能与专用硬件解耦。因此,虚拟网络功能(VNFs)可以分布到由不同网络位置的多个数据中心所拥有的共享和虚拟化平台上。通过虚拟链路相互连接的分布式VNFs的体系结构称为SFC (Service Function Chain)。在提高灵活性的同时,SFC有可能大幅降低网络服务的开放和运营成本。然而,SFC在链接数据中心上的可用性总是不如在单个数据中心上运行的网络功能,因为链上任何数据中心的故障都可能影响其可用性。为了提高sfc的可用性,本文提出了一种本地重路由策略,使用本地重路由路径(lrp)绕过sfc上故障的数据中心。此外,我们制定了一个优化模型,以最小化在网络中部署sfc和lrp时链路上的最大负载。因此,我们减少了由本地重路由策略引起的链路拥塞的潜在风险。然后,我们提出了一种随机四舍五入近似算法来解决优化问题,保持模型的竞争比为O(logn)。为了提高效率,我们还提出了一种快速的启发式算法。大量的仿真结果表明,所提出的算法可以提供高可用性的sfc,并且链路负载更加均衡。
Placement of Highly Available Virtual Network Functions Through Local Rerouting
The recent development of network function virtualization decouples network functions from dedicated hardware. Thus, virtual network functions (VNFs) can be distributed onto shared and virtualized platforms held by multiple data centers in various network locations. The architecture of the distributed VNFs interconnected by virtual links is denoted as the Service Function Chain (SFC). SFC has the potential to significantly reduce the opening and operating cost of the network services while improving the flexibility. However, the availability of SFC on chained up data centers is always inferior to that of network functions running on a single data center because the failure in any data center on the chain may affect its availability. To improve the availabilities of SFCs, in this paper we propose a local rerouting strategy to bypass the failed data centers on SFCs using locally rerouted paths (LRPs). Furthermore, we formulate an optimization model to minimize the maximum load on links while deploying SFCs and LRPs into the network. Thus, we reduce the potential risk of congested links caused by the local rerouting strategy. We then propose a randomized rounding approximation algorithm to solve the optimization problem, preserving the competitive ratio of O(logn) for the model. We also propose a fast heuristic algorithm to improve the efficiency. Our extensive simulation results show that the proposed algorithms can provide highly available SFCs with more balanced link load.