当前位置: X-MOL 学术J. Syst. Archit. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A novel frame preemption model in TSN networks
Journal of Systems Architecture ( IF 3.7 ) Pub Date : 2021-01-27 , DOI: 10.1016/j.sysarc.2021.102037
Mohammad Ashjaei , Mikael Sjödin , Saad Mubeen

This paper identifies a limitation in the frame preemption model in the TSN standard (IEEE 802.1Q-2018), due to which high priority frames can experience significantly long blocking delays, thereby exacerbating their worst-case response times. This limitation can have a considerable impact on the design, analysis and performance of TSN-based systems. To address this limitation, the paper presents a novel and more efficient frame preemption model in the TSN standard that allows over 90% reduction in the maximum blocking delay leading to lower worst-case response times of high priority frames compared to the frame preemption model used in the existing works. The paper also shows that the improvement becomes even more significant in multi-switch TSN networks. In order to evaluate the effects of preemption, the paper performs simulations by enabling and disabling preemptions as well as enabling and disabling the Hold/Release mechanism supported by TSN. Furthermore, the paper performs a comparative evaluation of the two models of frame preemption in TSN using simulations. The evaluation shows that the maximum response times of high priority frames can be significantly reduced with very small impact on the response times of lower priority frames. The paper also shows the improvement in the maximum response times of higher priority frames using an automotive industrial use case that employs a multi-hop TSN network for on-board communication.



中文翻译:

TSN网络中一种新颖的帧抢占模型

本文确定了TSN标准(IEEE 802.1Q-2018)中的帧抢占模型中的限制,由于该限制,高优先级的帧可能会经历相当长的阻塞延迟,从而加剧了它们在最坏情况下的响应时间。此限制可能对基于TSN的系统的设计,分析和性能产生重大影响。为了解决此限制,本文提出了一种新颖且更有效的TSN标准帧抢占模型,与使用的帧抢占模型相比,该模型允许最大阻塞延迟减少90%以上,从而降低了高优先级帧的最坏情况响应时间在现有作品中。本文还表明,这种改进在多交换TSN网络中变得更加重要。为了评估抢占的效果,本文通过启用和禁用抢占以及启用和禁用TSN支持的保留/释放机制来执行仿真。此外,本文使用仿真对TSN中两种帧抢占模型进行了比较评估。评估表明,高优先级帧的最大响应时间可以显着减少,而对低优先级帧的响应时间的影响很小。本文还显示了使用汽车工业用例的高优先级帧的最大响应时间的改进,该工业用例采用多跳TSN网络进行车载通信。本文利用仿真对TSN中两种帧抢占模型进行了比较评估。评估表明,高优先级帧的最大响应时间可以显着减少,而对低优先级帧的响应时间的影响很小。本文还显示了使用汽车工业用例的高优先级帧的最大响应时间的改进,该工业用例采用多跳TSN网络进行车载通信。本文利用仿真对TSN中两种帧抢占模型进行了比较评估。评估表明,高优先级帧的最大响应时间可以显着减少,而对低优先级帧的响应时间的影响很小。本文还显示了使用汽车工业用例的高优先级帧的最大响应时间的改进,该工业用例采用多跳TSN网络进行车载通信。

更新日期:2021-02-01
down
wechat
bug