{"title":"NFV中可用性感知的SFC嵌入:一种聚类方法","authors":"Pegah Torkamandi, S. Khorsandi, Bahador Bakhshi","doi":"10.1109/IranianCEE.2019.8786672","DOIUrl":null,"url":null,"abstract":"Network Function Virtualization (NFV) was recently proposed to improve the flexibility of network service provisioning and reduce the deployment costs by the decoupling the traditional network functions from the dedicated proprietary hardware equipment and turning them into the software-based virtualized entity, which is called a Virtual Network Function (VNF). Typically, the VNFs are concatenated together in a sequential order to form a Service Function Chains (SFCs) which provide specific services for users. Since failure of each VNF leads to breaking down the entire service chain, it is critical that service providers offer guaranteed and highly available SFCs against the failures. Previous works for simplicity assumed a completely available network infrastructure or just one failure in a same time; further, they increase availability of a SFC by adding dedicated backups which is costly. Unlike previous works, we investigate the problem of guarantying the required availability of SFCs against the multiple failures in a same time with minimum resource consumption. We proposed a new Availability-aware Clustered SFC Embedding (AC-SFC) algorithm which not only satisfies the availability requirement of SFC but also reduces the footprints of backup resource by using the Share Protection Cluster (SPC). Further, this problem is formulated as an Integer Linear Programming (ILP). We proposed a heuristic algorithm to overcome the ILP's complexity and new technique for calculating the availability of SPC. The numerical results show our heuristic algorithm works efficiency and reduces the network resource consumption in different network infrastructure topologies.","PeriodicalId":6683,"journal":{"name":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","volume":"19 1","pages":"1922-1928"},"PeriodicalIF":0.0000,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Availability Aware SFC Embedding in NFV: A Clustering Approach\",\"authors\":\"Pegah Torkamandi, S. Khorsandi, Bahador Bakhshi\",\"doi\":\"10.1109/IranianCEE.2019.8786672\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Network Function Virtualization (NFV) was recently proposed to improve the flexibility of network service provisioning and reduce the deployment costs by the decoupling the traditional network functions from the dedicated proprietary hardware equipment and turning them into the software-based virtualized entity, which is called a Virtual Network Function (VNF). Typically, the VNFs are concatenated together in a sequential order to form a Service Function Chains (SFCs) which provide specific services for users. Since failure of each VNF leads to breaking down the entire service chain, it is critical that service providers offer guaranteed and highly available SFCs against the failures. Previous works for simplicity assumed a completely available network infrastructure or just one failure in a same time; further, they increase availability of a SFC by adding dedicated backups which is costly. Unlike previous works, we investigate the problem of guarantying the required availability of SFCs against the multiple failures in a same time with minimum resource consumption. We proposed a new Availability-aware Clustered SFC Embedding (AC-SFC) algorithm which not only satisfies the availability requirement of SFC but also reduces the footprints of backup resource by using the Share Protection Cluster (SPC). Further, this problem is formulated as an Integer Linear Programming (ILP). We proposed a heuristic algorithm to overcome the ILP's complexity and new technique for calculating the availability of SPC. The numerical results show our heuristic algorithm works efficiency and reduces the network resource consumption in different network infrastructure topologies.\",\"PeriodicalId\":6683,\"journal\":{\"name\":\"2019 27th Iranian Conference on Electrical Engineering (ICEE)\",\"volume\":\"19 1\",\"pages\":\"1922-1928\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 27th Iranian Conference on Electrical Engineering (ICEE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/IranianCEE.2019.8786672\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 27th Iranian Conference on Electrical Engineering (ICEE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IranianCEE.2019.8786672","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
摘要
网络功能虚拟化(Network Function Virtualization, NFV)是近年来提出的一种网络功能虚拟化技术,它将传统的网络功能从专用的专有硬件设备中分离出来,并将其转化为基于软件的虚拟实体,即虚拟网络功能(Virtual Network Function, VNF),以提高网络服务提供的灵活性和降低部署成本。通常情况下,将多个VNFs按顺序连接在一起,形成一个sfc (Service Function Chains),为用户提供特定的服务。由于每个VNF的故障都会导致整个服务链的中断,因此服务提供商提供有保证的高可用性sfc以应对故障是至关重要的。为了简单起见,之前的工作假设了一个完全可用的网络基础设施,或者在同一时间只有一个故障;此外,它们通过添加昂贵的专用备份来提高SFC的可用性。与以前的工作不同,我们研究的问题是保证sfc在同一时间以最小的资源消耗应对多个故障所需的可用性。提出了一种新的可用性感知集群SFC嵌入(AC-SFC)算法,该算法既满足SFC的可用性要求,又利用共享保护集群(SPC)减少了备份资源的占用。进一步,将该问题表述为整数线性规划(ILP)。我们提出了一种启发式算法来克服ILP的复杂性,并提出了一种计算SPC可用性的新技术。数值结果表明,在不同的网络基础结构拓扑下,启发式算法能够有效地降低网络资源的消耗。
Availability Aware SFC Embedding in NFV: A Clustering Approach
Network Function Virtualization (NFV) was recently proposed to improve the flexibility of network service provisioning and reduce the deployment costs by the decoupling the traditional network functions from the dedicated proprietary hardware equipment and turning them into the software-based virtualized entity, which is called a Virtual Network Function (VNF). Typically, the VNFs are concatenated together in a sequential order to form a Service Function Chains (SFCs) which provide specific services for users. Since failure of each VNF leads to breaking down the entire service chain, it is critical that service providers offer guaranteed and highly available SFCs against the failures. Previous works for simplicity assumed a completely available network infrastructure or just one failure in a same time; further, they increase availability of a SFC by adding dedicated backups which is costly. Unlike previous works, we investigate the problem of guarantying the required availability of SFCs against the multiple failures in a same time with minimum resource consumption. We proposed a new Availability-aware Clustered SFC Embedding (AC-SFC) algorithm which not only satisfies the availability requirement of SFC but also reduces the footprints of backup resource by using the Share Protection Cluster (SPC). Further, this problem is formulated as an Integer Linear Programming (ILP). We proposed a heuristic algorithm to overcome the ILP's complexity and new technique for calculating the availability of SPC. The numerical results show our heuristic algorithm works efficiency and reduces the network resource consumption in different network infrastructure topologies.