基于高性能包处理架构VPP的带内网络遥测系统

doi:10.11959/j.issn.1000-436x.2021016

通信学报 ›› 2021, Vol. 42 ›› Issue (3): 75-90.doi: 10.11959/j.issn.1000-436x.2021016

基于高性能包处理架构VPP的带内网络遥测系统

潘恬¹, 林兴晨¹, 张娇¹, 黄韬¹^,², 刘韵洁¹^,²

¹ 北京邮电大学网络与交换技术国家重点实验室，北京 100876
² 网络通信与安全紫金山实验室，江苏南京 211111

修回日期:2020-10-26 出版日期:2021-03-25 发布日期:2021-03-01
作者简介:潘恬（1987- ），男，江苏常熟人，博士，北京邮电大学副教授、硕士生导师，主要研究方向为高速可编程网络、数据中心网络、低轨卫星网络等。
林兴晨（1995- ），男，福建宁德人，北京邮电大学硕士生，主要研究方向为高速可编程网络、数据中心网络等。
张娇（1986- ），女，河北保定人，博士，北京邮电大学副教授、博士生导师，主要研究方向为数据中心网络、传输控制协议、网络功能虚拟化等。
黄韬（1980- ），男，重庆人，博士，北京邮电大学教授、博士生导师，主要研究方向为未来网络体系架构、软件定义网络、信息中心网络等。
刘韵洁（1943- ），男，山东烟台人，中国工程院院士，主要研究方向为未来网络体系架构、网络操作系统、网络融合与演进、软件定义网络等。
基金资助:
国家重点研发计划基金资助项目(2019YFB1802600);国家自然科学基金资助项目(61702049)

In-band network telemetry system based on high-performance packet processing architecture VPP

Tian PAN¹, Xingchen LIN¹, Jiao ZHANG¹, Tao HUANG¹^,², Yunjie LIU¹^,²

¹ State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China
² Purple Mountain Laboratories, Nanjing 211111, China

Revised:2020-10-26 Online:2021-03-25 Published:2021-03-01
Supported by:
The National Key Research and Development Program of China(2019YFB1802600);The National Natural Science Foundation of China(61702049)

摘要/Abstract

摘要：

面向高性能虚拟网络转发架构及设备VPP，通过重构数据平面的流水线处理模块，设计了基于矢量数据包处理技术的带内网络遥测方案，并在此基础上通过引入源路由机制引导遥测报文走向，实现了一种带内全网遥测的扩展案例。最后，基于虚拟机组网方式，搭建网络拓扑并进行网络性能评测。实验结果表明，该遥测系统能够在0.13 ms精度的遥测时间间隔下覆盖被测网络链路，并能以较小代价实时监测虚拟网络的链路拥塞情况。鉴于虚拟网络设备已在数据中心内被广泛部署，该方案通过高精度的虚拟网络测量覆盖，有望为数据中心内多租户网络和网络功能虚拟化的大规模部署提供可靠性保障。

关键词: 带内网络遥测, 虚拟网络设备, 矢量包处理, 可编程数据平面

Abstract:

An in-band network telemetry-based system was built based on VPP, a high-performance virtual network forwarding architecture/device, via reorganizing the data plane pipeline processing modules.Moreover, a network-wide telemetry mechanism was further developed via embedding source routing into the probe packet header for specifying the route of probe packets.Finally, a virtual network topology was built with performance evaluation conducted.The evaluation shows that the telemetry system can monitor the network in a high precision of every 0.13 ms, detecting link congestion in real time with minor performance overhead.Given virtual network devices have been widely deployed in data centers, the proposed scheme is expected to improve the reliability of multi-tenancy and network function virtualization in data centers via high-precision, network-wide virtual link telemetry coverage.

Key words: in-band network telemetry, virtual network device, vector packet processing, programmable data plane

中图分类号:

TN92

潘恬, 林兴晨, 张娇, 黄韬, 刘韵洁. 基于高性能包处理架构VPP的带内网络遥测系统[J]. 通信学报, 2021, 42(3): 75-90.

Tian PAN, Xingchen LIN, Jiao ZHANG, Tao HUANG, Yunjie LIU. In-band network telemetry system based on high-performance packet processing architecture VPP[J]. Journal on Communications, 2021, 42(3): 75-90.

图/表 27

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

图15

图16

图17

图18

图19

图20

图21

表1

图22

图23

图24

图25

表2

参考文献 33

[1]	BARI M F , BOUTABA R , ESTEVES R ,et al. Data center network virtualization:a survey[J]. IEEE Communications Surveys ＆ Tutorials, 2012,15(2): 909-928.
[2]	DANIELS J . Server virtualization architecture and implementation[J]. XRDS:Crossroads,The ACM Magazine for Students, 2009,16(1): 8-12.
[3]	NELSON M , LIM B H , HUTCHINS G . Fast transparent migration for virtual machines[C]// USENIX Annual Technical Conference,General Track. Berkeley:USENIX Association, 2005: 391-394.
[4]	EGI N , GREENHALGH A , HANDLEY M ,et al. Towards high performance virtual routers on commodity hardware[C]// Proceedings of the 2008 ACM CoNEXT Conference. New York:ACM Press, 2008: 1-12.
[5]	PFAFF B , PETTIT J , KOPONEN T ,et al. The design and implementation of open vswitch[C]// 12th USENIX Symposium on Networked Systems Design and Implementation. Berkeley:USENIX Association, 2015: 117-130.
[6]	BARACH D , LINGUAGLOSSA L , MARION D ,et al. High-speed software data plane via vectorized packet processing[J]. IEEE Communications Magazine, 2018,56(12): 97-103.
[7]	JIA C , PAN T , BIAN Z ,et al. Rapid detection and localization of gray failures in data centers via in-band network telemetry[C]// NOMS 2020-2020 IEEE/IFIP Network Operations and Management Symposium. Piscataway:IEEE Press, 2020: 1-9.
[8]	刘争争, 毕军, 周禹 ,等. 基于P4的主动网络遥测机制[J]. 通信学报, 2018,39(Z1): 162-169.
	LIU Z Z , BI J , ZHOU Y ,et al. Paradigm for proactive telemetry based on P4[J]. Journal on Communications, 2018,39(Z1): 162-169.
[9]	BOSSHART P , GIBB G , KIM H S ,et al. Forwarding metamorphosis:Fast programmable match-action processing in hardware for SDN[J]. ACM SIGCOMM Computer Communication Review, 2013,43(4): 99-110.
[10]	BOSSHART P , DALY D , GIBB G ,et al. P4:programming protocol-independent packet processors[J]. ACM SIGCOMM Computer Communication Review, 2014,44(3): 87-95.
[11]	KIM C , SIVARAMAN A , KATTA N ,et al. In-band network telemetry via programmable dataplanes[C]// Proceedings of the Symposium on SDN Research. New York:ACM Press, 2015: 1-2.
[12]	SONG E , PAN T , JIA C ,et al. INT-filter:mitigating data collection overhead for high-resolution in-band network telemetry[C]// GLOBECOM 2020-2020 IEEE Global Communications Conference. Piscataway:IEEE Press, 2020: 1-6.
[13]	CASE J D , FEDOR M , SCHOFFSTALL M L ,et al. RFC1157:simple network management protocol (SNMP)[R]. IETF, 1990-05.
[14]	BREITGAND D , RAZ D , SHAVITT Y . SNMP GetPrev:an efficient way to browse large MIB tables[J]. IEEE Journal on Selected Areas in Communications, 2002,20(4): 656-667.
[15]	CHUN J K , CHO K Y , CHO S H ,et al. Network management based on PC communication platform with SNMP and mobile agents[C]// Proceedings of 22nd International Conference on Distributed Computing Systems Workshops. Piscataway:IEEE Press, 2002: 222-227.
[16]	ANDREY L , FESTOR O , LAHMADI A ,et al. Survey of SNMP performance analysis studies[J]. International Journal of Network Management, 2009,19(6): 527-548.
[17]	PRAS A , DREVERS T , VAN DE MEENT R ,et al. Comparing the performance of SNMP and Web services-based management[J]. IEEE Transactions on Network and Service Management, 2004,1(2): 72-82.
[18]	ESTAN C , KEYS K , MOORE D ,et al. Building a better NetFlow[J]. ACM SIGCOMM Computer Communication Review, 2004,34(4): 245-256.
[19]	PHAAL P , PANCHEN S , MCKEE N . RFC3176:InMon corporation’s sFlow:a method for monitoring traffic in switched and routed networks[R]. IETF, 2001-09.
[20]	QUITTEK J , ZSEBY T , CLAISE B ,et al. Requirements for IP flow information export (IPFIX)[R]. RFC 3917 (informational),IETF, 2004-10.
[21]	马云龙, 张千里, 王继龙 . 基于 IPFIX 的网络流量日志系统[J]. 通信学报, 2013,34(Z2): 5-8.
	MA Y L , ZHANG Q L , WANG J L . Network traffic analysis system based on IPFIX[J]. Journal on Communications, 2013,34(Z2): 5-8.
[22]	GUO C , YUAN L , XIANG D ,et al. Pingmesh:a large-scale system for data center network latency measurement and analysis[C]// Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. New York:ACM Press, 2015: 139-152.
[23]	LI Y , MIAO R , LIU H H ,et al. HPCC:high precision congestion control[C]// Proceedings of the ACM Special Interest Group on Data Communication. New York:ACM Press, 2019: 44-58.
[24]	KATTA N , HIRA M , KIM C ,et al. Hula:scalable load balancing using programmable data planes[C]// Proceedings of the Symposium on SDN Research. Piscataway:IEEE Press, 2016: 1-12.
[25]	JIN X , LI X , ZHANG H ,et al. Netcache:balancing key-value stores with fast in-network caching[C]// Proceedings of the 26th Symposium on Operating Systems Principles. Piscataway:IEEE Press, 2017: 121-136.
[26]	MIAO R , ZENG H , KIM C ,et al. Silkroad:making stateful layer-4 load balancing fast and cheap using switching ASICs[C]// Proceedings of the Conference of the ACM Special Interest Group on Data Communication. New York:ACM Press, 2017: 15-28.
[27]	The P4. org Applications Working Group.In-band network telemetry (int) dataplane specification (version 1.0)[R].(2018-04)[2020-08-04].
[28]	Huawei. World’s first iFIT pilot on the 5G transport network successfully completed by beijing Unicom and Huawei[R].(2019-06-28)[2020-08-04].
[29]	Broadcom. Broadcom trident 4 delivers disruptive economics for enterprise data center and campus networks[R].(2019-06-11)[2020-08-04].
[30]	PANTELOPOULOS A . Towards accurate simulations of programmable dataplanes[D]. Zurich:ETH Zurich, 2017.
[31]	SUNSHINE C A . Source routing in computer networks[J]. ACM SIGCOMM Computer Communication Review, 1977,7(1): 29-33.
[32]	PAN T , SONG E , BIAN Z ,et al. INT-path:towards optimal path planning for in-band network-wide telemetry[C]// IEEE INFOCOM 2019-IEEE Conference on Computer Communications. Piscataway:IEEE Press, 2019: 487-495.
[33]	朱祖勍, 孔嘉伟, 牛彬 ,等. 基于深度学习的面向 IP-over-EON 的可编程跨层网络业务性能感知系统[J]. 通信学报, 2019,40(11): 171-179.
	ZHU Z Q , KONG J W , NIU B ,et al. DL-assisted programmable multilayer network application awareness system for IP-over-EON[J]. Journal on Communications, 2019,40(11): 171-179.

转发设备	吞吐率/(Mbit·s^-1)	时延/ms	丢包率
VPP	99.9	0.014	0.019%
OVS	98.6	0.038	0.038%
BMv2	75.6	10.8	23.4%

背景流量	100 kbit/s INT探测速率/μs	1 Mbit/s INT探测速率/μs	10 Mbit/s INT探测速率/μs
无背景流量(3~8 s)	9.7	6.1	15.9
90 Mbit/s背景流量(15~25 s)	13.6	10.8	52.9
200 Mbit/s背景流量(30~40 s)	48.1	41.8	137.9

基于高性能包处理架构VPP的带内网络遥测系统

In-band network telemetry system based on high-performance packet processing architecture VPP

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 27

参考文献 33

相关文章 2

Metrics

推荐阅读 0

[1]	黄韬, 刘江, 汪硕, 张晨, 刘韵洁. 未来网络技术与发展趋势综述[J]. 通信学报, 2021, 42(1): 130-150.
[2]	朱祖勍,孔嘉伟,牛彬,唐绍飞,房红强,刘思祺. 基于深度学习的面向IP-over-EON的可编程跨层网络业务性能感知系统[J]. 通信学报, 2019, 40(11): 171-179.