芯片间时间触发消息堆叠调度方法

臧光界; 李峭; 王彤; 熊华钢

doi:10.13700/j.bh.1001-5965.2021.0553

芯片间时间触发消息堆叠调度方法

doi: 10.13700/j.bh.1001-5965.2021.0553

臧光界^{1, 2},
李峭^2, ,,
王彤²,
熊华钢²

1.
中国航空研究院，北京 100029
2.
北京航空航天大学电子信息工程学院，北京 100191

基金项目: 国家自然科学基金(62071023)

详细信息

通讯作者:
E-mail：avionics@buaa.edu.cn

中图分类号: TP393
计量
- 文章访问数: 180
- HTML全文浏览量: 38
- PDF下载量: 7
- 被引次数: 0
出版历程
- 收稿日期: 2021-09-15
- 录用日期: 2022-01-16
- 网络出版日期: 2022-02-15
- 整期出版日期: 2023-07-31

Stacking scheduling method for time-triggered messages in off-chip network

ZANG Guangjie^{1, 2},
LI Qiao^{2
, ,},
WANG Tong²,
XIONG Huagang²

1.
Chinese Aeronautical Establishment，Beijing 100029，China
2.
School of Electronic Information Engineering，Beihang University ，Beijing 100191，China

Funds: National Natural Science Foundation of China (62071023)

More Information

Corresponding author: E-mail：avionics@buaa.edu.cn

摘要

摘要:
时间触发（TT）通信方式应用于芯片间互连网络，以保证航空电子通信任务之间消息传递的严格时间确定性。当航空电子任务具有多种操作模式，属于不同模式的芯片间的时间触发调度表会重叠占用时隙，提出芯片间时间触发消息堆叠调度方法，以提高利用网络资源的灵活性和效率，同时减小应用层消息由于等待时间触发时间窗的排队延迟。仿真实验表明：与超调度方法相比，所提方法能够减小芯片间互连网络中时间触发消息的总端到端延迟和链路平均时隙占用率，对于端到端延迟时间较长且链路平均承载消息传输较多的场景，采用所提方法减少端到端延迟的效果更显著。
- 芯片间互连 /
- 时间触发消息 /
- 操作模式 /
- 堆叠调度 /
- 端到端延迟
Abstract:
To ensure the strict time determinism of message transmission between avionics communication tasks, time-triggered (TT) communication methods are also applied to the off-chip interconnection network. When avionics tasks have multiple operation modes, the time-triggered schedules between chips belonging to different modes will overlap and occupy time slots. The stacking scheduling method for time-triggered messages in off-chip network, which can reduce the queuing delay of application layer messages due to the waiting TT time window, is proposed to improve the flexibility and efficiency of using network resources,. Simulation experiments show that compared with the super scheduling method, the stack scheduling method can reduce the total end-to-end delay of TT messages in the off-chip interconnection network and the average time slot occupancy rate of the link. For scenarios where the end-to-end delay time is long and the average link bearer message transmission is large, the effect of using the stack scheduling method to reduce the end-to-end delay is more significant.
- off-chip interconnection /
- time-triggered message /
- operation mode /
- stacking scheduling /
- end-to-end delay

HTML全文

图 1 芯片间互连结构示意

Figure 1. Sketch map of off-chip interconnection

下载: 全尺寸图片幻灯片

图 2 物理链路在多模式情况下使用超调度方法的调度结果

Figure 2. Scheduling results of physical link using super-schedule approach in multi-mode

下载: 全尺寸图片幻灯片

图 3 物理链路在多模式情况下使用堆叠调度方法的调度结果

Figure 3. Scheduling results of physical link using stacking-schedule approach in multi-mode

下载: 全尺寸图片幻灯片

图 4 芯片间互连网络模式切换过程实例

Figure 4. An example of mode-change in off-chip interconnection network

下载: 全尺寸图片幻灯片

图 5 不同条件下端到端延迟的变化

Figure 5. Change of end-to-end delay under different conditions

下载: 全尺寸图片幻灯片

图 6 不同条件下链路平均时隙占用率的变化

Figure 6. Changes of average slot utilization rate of link under different conditions

下载: 全尺寸图片幻灯片

图 7 TRS调度方法端到端延迟增加时堆叠调度方法端到端延迟增加趋势

Figure 7. Increase trend of end-to-end delay of stack scheduling method, as end-to-end delay in TRS increases

下载: 全尺寸图片幻灯片

图 8 堆叠调度方法端到端延迟减小量随平均链路传递消息数变化

Figure 8. Change of amount of end-to-end delay reduction in stack scheduling method varies with average number of messages delivered on link

下载: 全尺寸图片幻灯片

图 9 堆叠调度方法端到端延迟随操作模式数变化

Figure 9. Change of end-to-end delay in stack scheduling method varies with number of messages’ modes

下载: 全尺寸图片幻灯片

表 1 实验参数设置

Table 1. Experimental parameter settings

实验组号消息组数每组消息数网络芯片数操作模式数

1 19 10~150 10 3
2 8 50 5~15 5
3 8 50 10 3~10

下载: 导出CSV

参考文献(20)

[1]	WOLFIG R, JAKOVLJEVIC M. Distributed IMA and DO-297: Architectural, communication and certification attributes[C]// IEEE/AIAA 27th Digital Avionics Systems Conference. Piscataway: IEEE Press, 2008: 1. E. 4-1.
[2]	何锋. 机载网络技术基础[M]. 北京: 国防工业出版社, 2018: 9-15. HE F. Fundamentals of airborne network[M]. Beijing: National Defense Industry Press, 2018: 9-15 (in Chinese).
[3]	蒲小勃. 现代航空电子系统与综合[M]. 北京: 航空工业出版社, 2013: 70-86. PU X B. Modern avionics system and integration[M]. Beijing: Aviation Industry Press, 2013: 70-86(in Chinese).
[4]	AHMADIAN H, OBERMAISSER R, PEREZ J. Distributed real-time architecture for mixed criticality systems[M]. Boca Raton: Taylor & Francis Group, 2018: 51-57.
[5]	PAUKOVITS C, KOPETZ H. Concepts of switching in the time-triggered network-on-chip[C]//2008 14th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications. Piscataway: IEEE Press, 2008: 120-129.
[6]	KOPETZ H, STEINER W. Real-time systems: Design principles for distributed embedded applications[M]. 2nd ed. Berlin: Springer International Publishing, 2011: 326-337.
[7]	DURRIEU G, FOHLER G, GALA G, et al. Dreams about reconfiguration and adaptation in avionics[C]//Proceedings Embedded Real Time Software and Systems. Toulouse: HAL Science Ouverte, 2016: 48-57.
[8]	URBINA M, OBERMAISSER R. A gateway core between on-chip and off-chip networks for an AUTOSAR message-based multi-core platform[C]//Automotive Meets Electronics. Dortmund: VDE Verlag GmbH, 2016: 1-6.
[9]	OLIVER R S, CRACIUNAS S S. Hierarchical scheduling over off- and on-chip deterministic networks[J]. ACM SIGBED Review, 2016, 13(4): 14-19. doi: 10.1145/3015037.3015039
[10]	STEINER W. An evaluation of SMT-based schedule synthesis for time-triggered multi-hop networks[C]//Real-Time Systems Symposium. Piscataway: IEEE Press, 2011: 375-384.
[11]	CRACIUNAS S S, OLIVIER R S. SMT-based task- and network-level static schedule generation for time-triggered networked systems[C]//22nd International Conference on Real-Time Networks and Systems (RTNS 2014). New York: ACM, 2014: 45-54.
[12]	郑重, 何锋, 李浩若, 等. 基于贪婪随机自适应搜索法的TTE通信调度算法[J]. 北京航空航天大学学报, 2021, 47(11): 2268-2276. doi: 10.13700/j.bh.1001-5965.2020.0382 ZHENG Z, HE F, LI H R, et al. Scheduling algorithm of TTE network based on greedy randomized adaptive search procedure[J]. Journal of Beijing University of Aeronautics and Astronautics, 2021, 47(11): 2268-2276(in Chinese). doi: 10.13700/j.bh.1001-5965.2020.0382
[13]	孔韵雯, 李峭, 熊华钢, 等. 片间综合化互连时间触发通信调度方法[J]. 航空学报, 2018, 39(2): 258-267. KONG Y W, LI Q, XIONG H G, et al. Time-triggered communication scheduling method for off-chip integrated interconnection[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(2): 258-267(in Chinese).
[14]	汪晶晶, 李峭, 熊华钢, 等. 芯片间时间触发通信综合规划方法及其优化[J]. 北京航空航天大学学报, 2020, 46(1): 170-180. doi: 10.13700/j.bh.1001-5965.2019.0136 WANG J J, LI Q, XIONG H G, et al. Integrated planning method and optimization for off-chip time-triggered communication[J]. Journal of Beijing University of Aeronautics and Astronautics, 2020, 46(1): 170-180(in Chinese). doi: 10.13700/j.bh.1001-5965.2019.0136
[15]	MOUTINHO L, PEDREIRAS P, ALMEIDA L. A real-time software defined networking framework for next-generation industrial networks[J]. IEEE Access, 2019, 7: 164468-164479. doi: 10.1109/ACCESS.2019.2952242
[16]	CRACIUNAS S, SERNA OLIVER R S. Online incremental scheduling method for deterministic networks: US20170070439[P]. 2017-03-09.
[17]	HEILMANN F, SYED A, FOHLER G. Mode-changes in COTS time-triggered network hardware without online reconfiguration[J]. ACM SIGBED Review, 2016, 13(4): 55-60. doi: 10.1145/3015037.3015046
[18]	SYED A. Model-based design and adaptive scheduling of distributed real-time systems[D]. Kaiserslautern: Technical University of Kaiserslautern, 2018: 60-63.
[19]	CHEN J C, DU C L, XIE F, et al. Schedulability analysis of non-preemptive strictly periodic tasks in multi-core real-time systems[J]. Real-Time Systems, 2016, 52(3): 239-271. doi: 10.1007/s11241-015-9226-z
[20]	TANG X Q, LI Q, ZUO Y J, et al. Online schedule of sporadic life-critical traffic in TTEthernet[C]//2019 IEEE 5th International Conference on Computer and Communications (ICCC). Piscataway: IEEE Press, 2020: 1312-1316.