虚拟化SDN中延迟感知动态管理程序的放置和重新配置

Saquib Amjad, Amir Varasteh, N. Deric, C. M. Machuca
{"title":"虚拟化SDN中延迟感知动态管理程序的放置和重新配置","authors":"Saquib Amjad, Amir Varasteh, N. Deric, C. M. Machuca","doi":"10.1109/NoF52522.2021.9609929","DOIUrl":null,"url":null,"abstract":"Software defined networking (SDN) provides different functionality and resource sharing capabilities with the aid of virtualization. In virtualized SDN, multiple SDN tenants can bring their controllers and different functions in the same physical substrate. The SDN hypervisor provides the link between the physical network substrate and its SDN tenants. Distributed hypervisor architecture can handle scalability, virtualization, and reassignment better than a centralized hypervisor architecture. However, distributed hypervisors may require synchronization and load balancing. Due to the dynamic control plane traffic, some network elements may suffer from increase in delay. Controller placement problem (CPP) and hypervisor placement problem (HPP) have tackled the static placement and mapping strategies.In this paper, we design a dynamic hypervisor assignment and reconfiguration with load balancing using integer linear programming. The proposed model focuses on optimizing multiple objectives: control plane latency, processing latency, and load balancing for distributed hypervisors. We provide different heuristics for the model and then perform evaluation on two topologies- Abilene and EU-nobel. The results show the trade off between load balancing and minimizing latency, and comparison of the performance of heuristics and MILP model.","PeriodicalId":314720,"journal":{"name":"2021 12th International Conference on Network of the Future (NoF)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Delay-Aware Dynamic Hypervisor Placement and Reconfiguration in Virtualized SDN\",\"authors\":\"Saquib Amjad, Amir Varasteh, N. Deric, C. M. Machuca\",\"doi\":\"10.1109/NoF52522.2021.9609929\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Software defined networking (SDN) provides different functionality and resource sharing capabilities with the aid of virtualization. In virtualized SDN, multiple SDN tenants can bring their controllers and different functions in the same physical substrate. The SDN hypervisor provides the link between the physical network substrate and its SDN tenants. Distributed hypervisor architecture can handle scalability, virtualization, and reassignment better than a centralized hypervisor architecture. However, distributed hypervisors may require synchronization and load balancing. Due to the dynamic control plane traffic, some network elements may suffer from increase in delay. Controller placement problem (CPP) and hypervisor placement problem (HPP) have tackled the static placement and mapping strategies.In this paper, we design a dynamic hypervisor assignment and reconfiguration with load balancing using integer linear programming. The proposed model focuses on optimizing multiple objectives: control plane latency, processing latency, and load balancing for distributed hypervisors. We provide different heuristics for the model and then perform evaluation on two topologies- Abilene and EU-nobel. The results show the trade off between load balancing and minimizing latency, and comparison of the performance of heuristics and MILP model.\",\"PeriodicalId\":314720,\"journal\":{\"name\":\"2021 12th International Conference on Network of the Future (NoF)\",\"volume\":\"18 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-10-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 12th International Conference on Network of the Future (NoF)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/NoF52522.2021.9609929\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 12th International Conference on Network of the Future (NoF)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/NoF52522.2021.9609929","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

软件定义网络(SDN)在虚拟化的帮助下提供不同的功能和资源共享能力。在虚拟化SDN中,多个SDN租户可以将各自的控制器和不同的功能带到同一个物理基板中。SDN管理程序提供物理网络基板与其SDN租户之间的链接。分布式管理程序体系结构可以比集中式管理程序体系结构更好地处理可伸缩性、虚拟化和重新分配。但是,分布式管理程序可能需要同步和负载平衡。由于控制平面流量是动态的,可能会导致部分网元的时延增加。控制器放置问题(CPP)和管理程序放置问题(HPP)解决了静态放置和映射策略。本文采用整数线性规划的方法,设计了一个具有负载均衡的动态管理程序分配与重构。所提出的模型侧重于优化多个目标:分布式管理程序的控制平面延迟、处理延迟和负载平衡。我们为模型提供了不同的启发式方法,然后对两种拓扑- Abilene和EU-nobel进行了评估。结果显示了负载平衡和最小化延迟之间的权衡,并比较了启发式和MILP模型的性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Delay-Aware Dynamic Hypervisor Placement and Reconfiguration in Virtualized SDN
Software defined networking (SDN) provides different functionality and resource sharing capabilities with the aid of virtualization. In virtualized SDN, multiple SDN tenants can bring their controllers and different functions in the same physical substrate. The SDN hypervisor provides the link between the physical network substrate and its SDN tenants. Distributed hypervisor architecture can handle scalability, virtualization, and reassignment better than a centralized hypervisor architecture. However, distributed hypervisors may require synchronization and load balancing. Due to the dynamic control plane traffic, some network elements may suffer from increase in delay. Controller placement problem (CPP) and hypervisor placement problem (HPP) have tackled the static placement and mapping strategies.In this paper, we design a dynamic hypervisor assignment and reconfiguration with load balancing using integer linear programming. The proposed model focuses on optimizing multiple objectives: control plane latency, processing latency, and load balancing for distributed hypervisors. We provide different heuristics for the model and then perform evaluation on two topologies- Abilene and EU-nobel. The results show the trade off between load balancing and minimizing latency, and comparison of the performance of heuristics and MILP model.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信