HSTC：TSN中的混合流量调度机制

尹长川; 李妍珏; 朱海龙; 何欣欣; 韩文璇

doi:10.11959/j.issn.1000-436x.2022103

您当前的位置：

首页 >

文章列表页 >

HSTC：TSN中的混合流量调度机制

学术论文 | 更新时间：2024-06-06

- HSTC：TSN中的混合流量调度机制
- HSTC: hybrid traffic scheduling mechanism in time-sensitive networking
- 通信学报 2022年43卷第6期页码：119-132
- 作者机构：
  
  1. 北京邮电大学先进信息网络北京实验室，北京 100876
  2. 北京邮电大学网络与交换技术国家重点实验室，北京 100876
- 作者简介：
  
  [ "尹长川（1968- ），男，山东潍坊人，博士，北京邮电大学教授、博士生导师，主要研究方向为无线通信系统与网络理论、统计信号处理、机器学习及其在无线通信中的应用、物联网技术等" ]
  [ "李妍珏（1997- ），女，云南玉溪人，北京邮电大学硕士生，主要研究方向为工业互联网、确定性网络、时间敏感网络等" ]
  [ "朱海龙（1987- ），男，山东菏泽人，博士，北京邮电大学讲师，主要研究方向为工业互联网、确定性网络、工业以太网、软件定义网络、时间敏感网络和车载网络等" ]
  [ "何欣欣（1987- ），女，江苏盐城人，博士，北京邮电大学讲师、硕士生导师，主要研究方向为新一代无线通信技术、车联网、时间敏感网络等" ]
  [ "韩文璇（1998- ），女，陕西西安人，北京邮电大学硕士生，主要研究方向为工业互联网、时间敏感网络和车载网络等" ]
- 基金信息：
  
  国家重点研发计划基金资助项目(2020YFB1805302);国家自然科学基金资助项目(61629101);国家自然科学基金资助项目(61671086)
- DOI：10.11959/j.issn.1000-436x.2022103
  中图分类号： TP393
- 网络首发：2022-06，
  
  纸质出版：2022-06-25
- 稿件说明：
移动端阅览
尹长川, 李妍珏, 朱海龙, 等. HSTC：TSN中的混合流量调度机制[J]. 通信学报, 2022,43(6):119-132.

Changchuan YIN, Yanjue LI, Hailong ZHU, et al. HSTC: hybrid traffic scheduling mechanism in time-sensitive networking[J]. Journal on Communications, 2022, 43(6): 119-132.
尹长川, 李妍珏, 朱海龙, 等. HSTC：TSN中的混合流量调度机制[J]. 通信学报, 2022,43(6):119-132. DOI： 10.11959/j.issn.1000-436x.2022103.

Changchuan YIN, Yanjue LI, Hailong ZHU, et al. HSTC: hybrid traffic scheduling mechanism in time-sensitive networking[J]. Journal on Communications, 2022, 43(6): 119-132. DOI： 10.11959/j.issn.1000-436x.2022103.

摘要

目的：随着工业4.0时代的到来，工业生产控制系统智能化程度越来越高，对信息的实时性和确定性传输提出了更高要求，时间敏感网络（Time-Sensitive Networking

TSN）由于具有良好的兼容性、低时延抖动等特性，被引入工业网络。为借助 TSN 实现工业网络中混合流高效传输，本文对TSN中的新型流量调度机制进行探索。

方法：研究新型 TSN 系统设计方案，基于集中式软件定义网络（SDN

software defined network）架构，设计了一种确定网络最小调度时隙的方法，并基于该最小时隙调整预订流（ST

scheduled traffic）类流量的采样周期，通过降低ST流对发送带宽的占用，为流预留（SR

stream reservation）类流量预留更多传输资源，从而提升网络可调度性。进一步针对SR流，提出奇偶映射的流映射策略，并且，当网络中出现SR流不可调度的情况时，设计流偏移规划（FOP

flow offset planning）算法，对SR流的注入时间进行偏移调整，通过提高系统资源利用率，进一步提升网络可调度性。

结果：为验证所提HSTC机制核心算法的性能，本文搭建实验平台，从ST流带宽占用、流调度优先级、SR 流映射、注入时隙选择等角度，将本文所提机制与现有机制进行性能仿真对比验证。实验设置网络最大帧长MTU=1500B，交换机单队列的最大缓存值BufSize=6MTU，链路速率1000 Mbit/s。每条ST流的最大采样周期和包长分别从集合{0.6

0.8

1.2

1.6}ms

{0.4

0.5

0.6

0.7

0.8

0.9

1}kB中随机选取，所有ST流的最小采样周期均设为0.1 ms

ST流的截止时间为其实际采样周期。SR流采样周期和包长分别从集合{4

20}ms

{1.5

2.5

3.5

4.5}kB中随机选取，每条SR流的截止时间取

</mo><msubsup> <mi>T</mi> <mtext>j</mtext> <mi>R</mi> </msubsup> <mo stretchy="false">]</mo></mrow></math>

范围内一个随机整数值。主要实验结果为：①调整ST流采样周期：若不进行采样周期调整，当流数目达到12，发送ST流占据了80%以上的网络带宽，而采用本文所提方案调整采样周期后，ST流的带宽占用大幅降低。②流调度优先级：本文所

提方案具有最优性能，其余方案性能排序依次为最大包长优先、最短截止时间优先、最小采样周期优先，相对于次优的最大包长优先方案，本机制所采用的加权排序方案至多能将网络调度成功率提高0.52。③SR流映射：随着SR 流数量的增加，本文所提出的奇偶映射方案对提升网络调度成功率的影响越显著，与按截止时间映射方案的网络调度成功率的最大差值达到20%。④注入时隙选择：相较于次优的随机选择注入方案，HSTC中的时隙排序方案可将网络调度成功率的值至多提升0.77，极限带宽利用率可达88%。⑤综合性能对比：在本文所研究场景下，以相同网络配置及流参数，与两篇代表性的现有文献所提方案进行性能对比，实验结果表明，HSTC机制实现了降低求解复杂度与提升调度性能的双重优化。

结论：目前，如何在工业网络中充分发挥 TSN 精准的流量调度能力，为生产控制系统提供确定性和实时性保障，仍是TSN的一个研究重点。为此，本文提出一种HSTC混合流量调度机制，该机制将TAS与CQF这2种现有方案进行有机结合，并根据时间敏感流和大带宽流的流特性，为两类流制定不同的调度策略。实验结果表明，HSTC机制通过提升系统资源利用率显著提高了网络可调度性，实现了TSN混合流的高效调度。现有对TSN的网络规划多基于离线调度场景，但在实际工业网络中，还存在由事件触发的少量突发流量，这些流无固定参数，但对系统的正常运转有着重要影响，因此如何改进现有方案，使之同时支持突发流的混合传输，是本文下一步的研究方向。

Abstract

Objectives: With the arrival of the industry 4.0 era

the industrial production control system is becoming more and more intelligent

which puts forward higher requirements for real-time and deterministic transmission of information. Time-Sensitive Networking (TSN) has been introduced into industrial network because of its good compatibility and low delay jitter. To realize efficient transmission of mixed traffic in industrial network with the help of TSN

we explored a new traffic scheduling mechanism in TSN.

Methods: Based on architecture of the centralized software defined network (SDN)

we designed a method to determine the minimum scheduling slot of the network and adjusted sampling period of Scheduled Traffic (ST) based on the minimum slot. By reducing occupation of transmission bandwidth by ST traffic

more transmission resources were reserved for Stream Reservation (SR) traffic to improve network schedulability. Furthermore

for SR traffic

a parity mapping scheme was proposed. When SR flow was not schedulable

a flow offset planning (FOP) algorithm was designed to offset injection time of SR flow

which can further improve schedulability of the network by improving utilization of system resources.

Results: To verify performan

ce of core algorithm of HSTC mechanism

we built an experimental platform and compared performance of our mechanism with existing mechanisms from the perspectives of ST traffic bandwidth occupation

traffic scheduling priority

SR traffic mapping

injection slot selection

etc. In the experiment

the maximum transmission unit (MTU) of network was 1500B

the maximum buffer size of single queue in switch was 6MTU

and link rate was 1000 Mbit/s. The maximum sampling period and packet length of each ST flow were randomly selected from the set {0.6

0.8

1.2

1.6}ms

{0.4

0.5

0.6

0.7

0.8

0.9

1}kB. Minimum sampling period of ST flow was set to 0.1 ms. And the deadline (DDL) of ST flow was its actual sampling period. Sampling period and packet length of SR flow were randomly selected from set {4

20}ms

{1.5

2.5

3.5

4.5}kB. DDL of each SR flow was a random integer value within range of

</mo><msubsup> <mi>T</mi> <mtext>j</mtext> <mi>R</mi> </msubsup> <mo stretchy="false">]</mo></mrow></math>

. Main experimental results are showed as follows: ①Adjusting sampling period of ST traffic: if sampling period was not adjusted

when the number of flows reached 12

transmitting ST traffic occupied more than 80% of network bandwidth. However

after adjusting sampling period by our scheme

the bandwidth occupation of ST traffic was greatly reduced. ②Traffic scheduling priority: weighted ranking scheme proposed in this paper had the best performance. Performance ranking of other schemes was the maximum packet length first

the shortest deadline first

and the minimum sampling period first. Compared with the sub optimal scheme

weighted ranking scheme can improve scheduling success rate by up to 0.52. ③SR traffic mapping: as the number of SR flows increased

impact of parity mappin

g scheme on improving scheduling success rate became more significant. The maximum difference between scheduling success rate of our scheme and the scheme which mapped according to DDL was 20%. ④Compared with the suboptimal random slot injection scheme

the slot sequencing scheme in HSTC can increase network scheduling success rate by up to 0.77

and the limit bandwidth utilization can reach 88%.⑤Overall performance comparison: we compared with two mechanisms which were proposed in two representative existing literatures under the same simulation parameter configuration. The experimental results showed that HSTC mechanism realized dual optimization of reducing solution complexity and improving scheduling performance.

Conclusions: How to make full use of TSN's accurate flow scheduling capability to provide certainty and real-time guarantee for production control system is still a research focus of TSN. Therefore

we proposed a mixed traffic scheduling mechanism called HSTC

which combined two existing schemes of Time-Aware Shaper (TAS) and Cyclic Queuing and Forwarding (CQF) and formulated different scheduling strategies for time-sensitive traffic and large bandwidth traffic according to their characteristic. The experimental results showed that HSTC mechanism significantly improved network schedulability by improving system resource utilization

and it realized efficient scheduling of mixed traffic. Existing network schedule schemes for TSN are mostly based on off-line scheduling scenarios. However

in actual industrial network

there is still a small number of burst traffic triggered by events. Burst traffic has no fixed parameters but has an important impact on the normal operation of the system. Therefore

how to improve our current mechanism to support mixed transmission of burst traffic at the same time is our next research direction.

关键词

Keywords

references

IEEE . IEEE standard for Ethernet:IEEE Std 802.3-2015 (Revision of IEEE Std 802.3-2012) [S ] . 2016 .

DECOTIGNIE J D . Ethernet-based real-time and industrial communications [J ] . Proceedings of the IEEE , 2005 , 93 ( 6 ): 1102 - 1117 .

SCHEMM E . SERCOS to link with Ethernet for its third generation [J ] . Computing and Control Engineering , 2004 , 15 ( 2 ): 30 - 33 .

黄韬 , 汪硕 , 黄玉栋 , 等 . 确定性网络研究综述 [J ] . 通信学报 , 2019 , 40 ( 6 ): 160 - 176 .

HUANG T , WANG S , HUANG Y D , et al . Survey of the deterministic network [J ] . Journal on Communications , 2019 , 40 ( 6 ): 160 - 176 .

STEINER W , CRACIUNAS S S , OLIVER R S . Traffic planning for time-sensitive communication [J ] . IEEE Communications Standards Magazine , 2018 , 2 ( 2 ): 42 - 47 .

NASRALLAH A , THYAGATURU A S , ALHARBI Z , et al . Ultra-low latency (ULL) networks:the IEEE TSN and IETF DetNet standards and related 5G ULL research [J ] . IEEE Communications Surveys ＆Tutorials , 2019 , 21 ( 1 ): 88 - 145 .

IEEE . IEEE standard for local and metropolitan area networks - timing and synchronization for time-sensitive applications in bridged local area networks:IEEE Std 802.1AS-2011 [S ] . 2011 .

IEEE . IEEE standard for local and metropolitan area networks—bridges and bridged networks - amendment 25:enhancements for scheduled traffic:IEEE Std 802.1Qbv-2015 (amendment to IEEE Std 802.1Q-2014 as amended by IEEE Std 802.1Qca-2015,IEEE Std 802.1Qcd-2015,and IEEE Std 802.1Q-2014/Cor 1-2015) [S ] . 2015 .

IEEE . IEEE standard for local and metropolitan area networks–bridges and bridged networks-amendment 29:cyclic queuing and forwarding:IEEE 802.1Qch-2017 [S ] . 2017 .

IEEE . IEEE standard for local and metropolitan area networks——virtual bridged local area networks-amendment 12:forwarding and queuing enhancements for time-sensitive streams:IEEE Std 802.1Qav-2009 (Amendment to IEEE Std 802.1Q-2005) [S ] . 2009 .

IEEE . IEEE standard for local and metropolitan area networks——bridges and bridged networks-amendment 31:stream reservation protocol (SRP) enhancements and performance improvements:IEEE Std 802.1Qcc-2018 (Amendment to IEEE Std 802.1Q-2018 as amended by IEEE Std 802.1Qcp-2018) [S ] . 2018 .

ANSAH F , ABID M A , DE MEER H . Schedulability analysis and GCL computation for time-sensitive networks [C ] // Proceedings of 2019 IEEE 17th International Conference on Industrial Informatics . Piscataway:IEEE Press , 2019 : 926 - 932 .

MEHEDI H M , KHAN S , FENG H , et al . Improved end to end delay in CBS using data compression for time sensitive network [C ] // Proceedings of 2021 2nd Information Communication Technologies Conference (ICTC) . Piscataway:IEEE Press , 2021 : 172 - 177 .

YAN J L , QUAN W , JIANG X Y , et al . Injection time planning:making CQF practical in time-sensitive networking [C ] // Proceedings of IEEE INFOCOM 2020 - IEEE Conference on Computer Communications . Piscataway:IEEE Press , 2020 : 616 - 625 .

QUAN W , YAN J L , JIANG X Y , et al . On-line traffic scheduling optimization in IEEE 802.1Qch based time-sensitive networks [C ] // Proceedings of 2020 IEEE 22nd International Conference on High Performance Computing and Communications; IEEE 18th International Conference on Smart City; IEEE 6th International Conference on Data Science and Systems . Piscataway:IEEE Press , 2020 : 369 - 376 .

LI E S , HE F , LI Q , et al . Bandwidth allocation of stream-reservation traffic in TSN [J ] . IEEE Transactions on Network and Service Management , 2021 , 19 ( 1 ): 741 - 755 .

CAO Z P , LIU Q R , LIU D P , et al . Enhanced system design and scheduling strategy for switches in time-sensitive networking [J ] . IEEE Access , 2021 , 9 : 42621 - 42634 .

HUANG Y D , WANG S , WU B W , et al . TACQ:enabling zero-jitter for cyclic-queuing and forwarding in time-sensitive networks [C ] // Proceedings of IEEE International Conference on Communications . Piscataway:IEEE Press , 2021 : 1 - 6 .

ZHANG J L , XU Q M , LU X Z , et al . Coordinated data transmission in time-sensitive networking for mixed time-sensitive applications [C ] // Proceedings of the 46th Annual Conference of the IEEE Industrial Electronics Society . Piscataway:IEEE Press , 2020 : 3805 - 3810 .

莫益军 , 杨子涵 , 刘辉宇 , 等 . 面向大规模确定性网络的全局循环排队与转发机制 [J ] . 电信科学 , 2021 , 37 ( 10 ): 55 - 65 .

MO Y J , YANG Z H , LIU H Y , et al . Global cyclic queuing and forwarding mechanism for large-scale deterministic networks [J ] . Telecommunications Science , 2021 , 37 ( 10 ): 55 - 65 .

PENG G Y , WANG S , HUANG Y D , et al . Traffic shaping at the edge:enabling bounded latency for large-scale deterministic networks [C ] // Proceedings of 2021 IEEE International Conference on Communications Workshops . Piscataway:IEEE Press , 2021 : 1 - 6 .

LEONARDI L , BELLO L L , PATTI G . Performance assessment of the IEEE 802.1Qch in an automotive scenario [C ] // Proceedings of 2020 AEIT International Conference of Electrical and Electronic Technologies for Automotive (AEIT AUTOMOTIVE) . Piscataway:IEEE Press , 2020 : 1 - 6 .

ARZEN K E , CERVIN A , EKER J , et al . An introduction to control and scheduling co-design [C ] // Proceedings of the 39th IEEE Conference on Decision and Control . Piscataway:IEEE Press , 2000 : 4865 - 4870 .

DAI X T , BURNS A . Period adaptation of real-time control tasks with fixed-priority scheduling in cyber-physical systems [J ] . Journal of Systems Architecture , 2020 , 103 : 101691 .

DAI X T , CHANG W L , ZHAO S , et al . A dual-mode strategy for performance-maximisation and resource-efficient CPS design [J ] . ACM Transactions on Embedded Computing Systems , 2019 , 18 ( 5s ): 1 - 20 .

ZHENG Y , WANG S , YIN S W , et al . Mix-flow scheduling for concurrent multipath transmission in time-sensitive networking [C ] // Proceedings of 2021 IEEE International Conference on Communications Workshops . Piscataway:IEEE Press , 2021 : 1 - 6 .

张艮山 , 刘旭宁 . 一种多重约束下确保成功率的云工作流调度方法 [J ] . 计算机应用与软件 , 2021 , 38 ( 2 ): 278 - 284 , 317 .

ZHANG G S , LIU X N . A cloud workflow scheduling method ensuring success rate under multiple constraints [J ] . Computer Applications and Software , 2021 , 38 ( 2 ): 278 - 284 , 317 .

IEEE . IEEE standard for local and metropolitan area networks—audio video bridging (AVB) systems:IEEE Std 802.1BA-2011 [S ] . 2011 .

HUANG Y D , WANG S , FENG T , et al . Towards network-wide scheduling for cyclic traffic in IP-based deterministic networks [C ] // Proceedings of 2021 4th International Conference on Hot Information-Centric Networking . Piscataway:IEEE Press , 2021 : 117 - 122 .

浏览量

1813

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

时间确定性网络技术与应用研究进展

GAI使能网络智能化：基于LLM的网络操作系统

面向混合多域组网的软件定义时间敏感网络

基于网络演算的LEO卫星网络时延上界分析研究

基于帧保序缓存的时间敏感网络FRER机制