欧阳一鸣,陈荣景,王 奇,梁华国.基于故障感知容错路由算法的可重构架构设计[J].电子测量与仪器学报,2023,37(5):108-116
基于故障感知容错路由算法的可重构架构设计
Reconfigurable architecture design based on fault-aware fault-tolerant routing algorithm
  
DOI:
中文关键词:  片上网络  容错  可重构  路由算法
英文关键词:network-on-chip (NoC)  fault tolerant  reconfigurable  routing algorithm
基金项目:国家自然科学基金(61874157,61674048)项目资助
作者单位
欧阳一鸣 1. 合肥工业大学计算机与信息学院 
陈荣景 1. 合肥工业大学计算机与信息学院 
王 奇 2. 合肥工业大学电子科学与应用物理学院 
梁华国 2. 合肥工业大学电子科学与应用物理学院 
AuthorInstitution
Ouyang Yiming 1. School of Computer and Information, Hefei University of Technology 
Chen Rongjing 1. School of Computer and Information, Hefei University of Technology 
Wang Qi 2. School of Electronic Science and Applied Physics, Hefei University of Technology 
Liang Huaguo 2. School of Electronic Science and Applied Physics, Hefei University of Technology 
摘要点击次数: 189
全文下载次数: 475
中文摘要:
      当路由器发生永久性故障时会影响网络的通信性能,现有的容错方法大多都是采用重路由策略,绕行路径的不确定性 不仅会带来较长的绕行延迟甚至故障节点周围形成热点导致死锁。 针对二维网状网络中的各种故障情况,设计了一种新的路 由器架构———DRRA,通过添加的组件将不同的输入端口和输出端口连接起来,并定义了 3 种不同的具体连接方式,当数据包遇 到故障节点时,会根据具体故障位置及路由信息选择合适方式直接绕过该故障节点,保证网络的连接性。 实验结果表明,本文 所提出的方案与其他容错方案性比不会产生过多的硬件开销,并且在网络存在多个故障节点的情况下保持良好的性能和可靠 性。 在热点流量模式下,本文提出的方案与 ReRS 方案相比可以降低 57. 4%的平均数据包传输延迟,与 MiCoF 相比可以降低 38. 9%的平均数据包传输延迟。
英文摘要:
      Network communication performance will be affected when permanent fault occurs, most of the existing fault tolerant methods use re-routing strategies, the uncertainty of the bypass selection can bring longer delays or even form hotspots around the failed node leading to deadlocks. In this paper, a new router architecture, DRRA, is proposed to address various failure cases in 2D mesh NoC. The different input and output ports are connected by the added components and three different connect methods are defined, when packets encounter a faulty node, it will select the appropriate method to directly bypass the faulty node based on the specific fault location and routing information to ensure the connectivity of the network. The experimental results show that compared with other fault-tolerant schemes, the proposed method has good performance and reliability in the presence of multiple failed nodes in the network and does not bring excessive hardware overhead. In the hotspot traffic mode, the proposed scheme in this paper can reduce the average packet transmission delay by 57. 4% compared with the ReRS scheme and 38. 9% compared with MiCoF.
查看全文  查看/发表评论  下载PDF阅读器