Latency-Aware Traffic Grooming for Dynamic Service Chaining in Metro Networks

Leila Askari, F. Musumeci, M. Tornatore
{"title":"Latency-Aware Traffic Grooming for Dynamic Service Chaining in Metro Networks","authors":"Leila Askari, F. Musumeci, M. Tornatore","doi":"10.1109/ICC.2019.8761290","DOIUrl":null,"url":null,"abstract":"Optical metro networks are currently evolving in response to the new requirements of emerging 5G services. Network Function Virtualization (NFV) is being leveraged as a platform to dynamically provision these services on top of Virtual Network Functions (VNFs), and central offices in metro areas are being upgraded to host processing units that can host the needed to provision services with stringent-latency and high-bandwidth requirements closer to users (i.e., edge computing). By concatenating these VNFs in a specific order and route traffic among them, operators generate a so-called “Service Chain“(SC). Considering the fact that, new 5G services have bandwidth requirements typically with sub-wavelength granularity, traffic grooming is required to achieve efficient network resources utilization. Since grooming affects the end-to-end latency of provisioned services, we investigate how to perform latency-aware traffic grooming, and we propose an algorithm for dynamic SC provisioning, that considers the latency requirements of each SC to decide about if grooming shall be allowed at intermediate network nodes. Our proposed algorithm tries to minimize the blocked bandwidth as well as number of nodes to host VNFs in the network (NFV-nodes) considering the nodes computational capacity, links bandwidth and end-to-end latency constraints. Results obtained from numerical evaluation show that, our algorithm is able to reduce the number of NFV-nodes up to 50%, while keeping amount of blocked bandwidth below a specific threshold.","PeriodicalId":402732,"journal":{"name":"ICC 2019 - 2019 IEEE International Conference on Communications (ICC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ICC 2019 - 2019 IEEE International Conference on Communications (ICC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICC.2019.8761290","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 13

Abstract

Optical metro networks are currently evolving in response to the new requirements of emerging 5G services. Network Function Virtualization (NFV) is being leveraged as a platform to dynamically provision these services on top of Virtual Network Functions (VNFs), and central offices in metro areas are being upgraded to host processing units that can host the needed to provision services with stringent-latency and high-bandwidth requirements closer to users (i.e., edge computing). By concatenating these VNFs in a specific order and route traffic among them, operators generate a so-called “Service Chain“(SC). Considering the fact that, new 5G services have bandwidth requirements typically with sub-wavelength granularity, traffic grooming is required to achieve efficient network resources utilization. Since grooming affects the end-to-end latency of provisioned services, we investigate how to perform latency-aware traffic grooming, and we propose an algorithm for dynamic SC provisioning, that considers the latency requirements of each SC to decide about if grooming shall be allowed at intermediate network nodes. Our proposed algorithm tries to minimize the blocked bandwidth as well as number of nodes to host VNFs in the network (NFV-nodes) considering the nodes computational capacity, links bandwidth and end-to-end latency constraints. Results obtained from numerical evaluation show that, our algorithm is able to reduce the number of NFV-nodes up to 50%, while keeping amount of blocked bandwidth below a specific threshold.
城域网络中动态业务链的时延感知流量疏导
光城域网目前正在不断发展,以满足新兴5G业务的新需求。网络功能虚拟化(NFV)正被用作一个平台,在虚拟网络功能(VNFs)之上动态地提供这些服务,城域的中心办公室正在升级为主机处理单元,可以托管更靠近用户的、具有严格延迟和高带宽要求的服务(即边缘计算)。通过按特定顺序将这些vnf连接起来,并在它们之间路由流量,运营商生成了所谓的“服务链”(SC)。考虑到5G新业务的带宽需求通常为亚波长粒度,需要进行流量疏导,以实现高效的网络资源利用。由于疏导影响已配置服务的端到端延迟,我们研究了如何执行延迟感知流量疏导,并提出了一种动态SC配置算法,该算法考虑每个SC的延迟需求,以决定是否允许在中间网络节点进行疏导。我们提出的算法在考虑节点计算能力、链路带宽和端到端延迟约束的情况下,尽量减少阻塞带宽以及网络中承载VNFs的节点(nfv节点)的数量。数值评估结果表明,该算法能够将nfv节点数量减少50%,同时将阻塞带宽数量保持在特定阈值以下。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
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