arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.02040cs.NI
收藏

SD-FFR:自动化立体仓库中软件定义的快速Failure Recovery机制

SD-FFR: Software Defined Fast Failure Recovery Mechanism in the Automatic Warehouse

Ting-Cian Bai, Chin-Ya Huang

摘要:

由于物联网技术的快速发展,自动导引车(AGV)通过无线网络与工业控制系统(ICS)交互,以支持自动化仓库中的货物分配。然而,在无线网络链路故障和链路拥塞的情况下,AGV和ICS之间的消息交换将经历大的分组丢失或长的传输延迟。因此,仓库自动化的性能将降低。本文提出了软件定义的快速故障恢复(SD-FFR)机制,旨在提高网络的可靠性,避免由于负载不平衡引起的链路拥塞。利用SD-FFR,当链路故障发生时,SDN控制器在毫秒内主动检测链路故障,然后相应地重新路由受影响的业务流。所提出的SD-FFR机制还考虑了当新业务流加入或网络中发生链路故障时在选择路由路径时的负载平衡。评估结果表明,网络性能可以有效地提高时间。

英文摘要:

Due to the rapid development of IoT technology, automatic guided vehicles (AGVs) interact with an industrial control system (ICS) through the wireless network to support the freight distribution in the automated warehouse. However, the message exchange among AGVs and the ICS would experience large packet loss or long transmission delay, in the presence of wireless network link failure and link congestion. Therefore, the performance of warehouse automation would be degraded. In this paper, we propose the Software Defined Fast Failure Recovery(SD-FFR) mechanism, aiming to improve network reliability and avoid link congestion caused by load unbalance. With SD-FFR, when link failure occurs, the SDN controller actively detects the link failure within milliseconds, and then reroutes the affected traffic flows accordingly. The proposed SD-FFR mechanism also takes load balance into consideration in selecting routing paths when a new traffic flow joins or link failure occurs in the network. The evaluation results show that network performance can be time efficiently improved.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑