Towards a URLLC-Aware Programmable Data Path with P4 for Industrial 5G Networks

Kerim Gökarslan, Yagmur Sabucu Sandal, T. Tuğcu
{"title":"Towards a URLLC-Aware Programmable Data Path with P4 for Industrial 5G Networks","authors":"Kerim Gökarslan, Yagmur Sabucu Sandal, T. Tuğcu","doi":"10.1109/ICCWorkshops50388.2021.9473859","DOIUrl":null,"url":null,"abstract":"Having their own requirement and specifications, industrial networks mostly bank on traditionally more reliable wired technologies such as Ethernet and PROFINET. Recent developments in cellular technologies, more specifically 5G, bring a new era with ultra-reliable low-latency communication (URLLC) where networks can achieve six nines of reliability with latency values around 1 ms. Industries, thus, have a substantial interest in deploying 5G at factories as it can reduce both operational and capital costs while not compromising latency and reliability requirements. Unfortunately, today’s 5G networks are designed for the larger subscriber community in a city or country that has requirements significantly different from industrial networks. In this paper, we propose a novel programmable data path for industrial 5G networks in P4, a high-level programming language to control data plane in network devices, to achieve lower latency values while enabling network engineers to have a fine-grained real-time network monitoring and increasing network security using an in-network switch. Our design leverages the relaxations of operating industrial 5G networks compared to traditional multitenant cellular networks, such as the fact that a factory is both the network operator and user equipment (UE) operator. We implement our design in P416 with P4’s software switch BMV2 and demonstrate its benefits on Open5GS, an open-source C-based 5G core implementation, and UERANSIM, an open-source 5G UE and RAN simulator. Our thorough evaluations show that our design can reduce intra-cellular network latency up to 2x compared to the traditional 5G architecture. We further demonstrate that our system can enable network administrators to do fine-grained network monitoring at the ~10 ms polling interval rates without significantly affecting the existing traffic. Similarly, we demonstrate that security rules can be updated within 10 ms with a 95% confidence interval. Noting that we run experiments on a P4-based software switch, we expect to see much lower update intervals on a P4-hardware switch.","PeriodicalId":127186,"journal":{"name":"2021 IEEE International Conference on Communications Workshops (ICC Workshops)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"11","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE International Conference on Communications Workshops (ICC Workshops)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCWorkshops50388.2021.9473859","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 11

Abstract

Having their own requirement and specifications, industrial networks mostly bank on traditionally more reliable wired technologies such as Ethernet and PROFINET. Recent developments in cellular technologies, more specifically 5G, bring a new era with ultra-reliable low-latency communication (URLLC) where networks can achieve six nines of reliability with latency values around 1 ms. Industries, thus, have a substantial interest in deploying 5G at factories as it can reduce both operational and capital costs while not compromising latency and reliability requirements. Unfortunately, today’s 5G networks are designed for the larger subscriber community in a city or country that has requirements significantly different from industrial networks. In this paper, we propose a novel programmable data path for industrial 5G networks in P4, a high-level programming language to control data plane in network devices, to achieve lower latency values while enabling network engineers to have a fine-grained real-time network monitoring and increasing network security using an in-network switch. Our design leverages the relaxations of operating industrial 5G networks compared to traditional multitenant cellular networks, such as the fact that a factory is both the network operator and user equipment (UE) operator. We implement our design in P416 with P4’s software switch BMV2 and demonstrate its benefits on Open5GS, an open-source C-based 5G core implementation, and UERANSIM, an open-source 5G UE and RAN simulator. Our thorough evaluations show that our design can reduce intra-cellular network latency up to 2x compared to the traditional 5G architecture. We further demonstrate that our system can enable network administrators to do fine-grained network monitoring at the ~10 ms polling interval rates without significantly affecting the existing traffic. Similarly, we demonstrate that security rules can be updated within 10 ms with a 95% confidence interval. Noting that we run experiments on a P4-based software switch, we expect to see much lower update intervals on a P4-hardware switch.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
面向工业5G网络的基于P4的urllc感知可编程数据路径
工业网络有自己的需求和规范,主要依靠传统上更可靠的有线技术,如以太网和PROFINET。蜂窝技术的最新发展,更具体地说是5G,带来了一个超可靠低延迟通信(URLLC)的新时代,网络可以实现6倍的可靠性,延迟值约为1毫秒。因此,工业对在工厂部署5G非常感兴趣,因为它可以降低运营和资本成本,同时不影响延迟和可靠性要求。不幸的是,今天的5G网络是为城市或国家的更大用户社区设计的,其需求与工业网络有很大不同。在本文中,我们提出了一种新的可编程数据路径,用于工业5G网络的P4,这是一种高级编程语言,用于控制网络设备中的数据平面,以实现更低的延迟值,同时使网络工程师能够使用网内交换机进行细粒度的实时网络监控并提高网络安全性。与传统的多租户蜂窝网络相比,我们的设计充分利用了运营工业5G网络的便利性,例如工厂既是网络运营商又是用户设备(UE)运营商。我们用P4的软件交换机BMV2在P416上实现了我们的设计,并在Open5GS(一个开源的基于c的5G核心实现)和UERANSIM(一个开源的5G终端和RAN模拟器)上展示了它的优势。我们的全面评估表明,与传统5G架构相比,我们的设计可以将蜂窝内网络延迟减少至多2倍。我们进一步证明,我们的系统可以使网络管理员以~10 ms的轮询间隔率进行细粒度的网络监视,而不会对现有流量产生显著影响。类似地,我们证明了安全规则可以在10毫秒内以95%的置信区间更新。注意到我们在基于p4的软件交换机上运行实验,我们期望在p4硬件交换机上看到更低的更新间隔。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
BML: An Efficient and Versatile Tool for BGP Dataset Collection Efficient and Privacy-Preserving Contact Tracing System for Covid-19 using Blockchain MEC-Based Energy-Aware Distributed Feature Extraction for mHealth Applications with Strict Latency Requirements Distributed Multi-Agent Learning for Service Function Chain Partial Offloading at the Edge A Deep Neural Network Based Environment Sensing in the Presence of Jammers
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1