【24h】

A provably good performance centric NoC topology

机译:以性能为中心的可证明的NoC拓扑

获取原文

摘要

As chip density increases rapidly with every process generation, the use of Network-on-Chip (NoC) has become the prevalent architecture for SoC, MPSoC, and, large scale CMP (Chip Multi Processor) based designs. Diverse NoC solutions have been proposed by the researchers to meet the enhanced on-chip communication requirements. Here, underlying network interconnection architecture (topology), router design and routing policy play an important role for overall system performance improvement. In this work, a 2D Hybrid Mesh based Star topology has been proposed with an objective of providing low latency, low channel contention and higher throughput based system. The observe experimental results show a maximum latency benefit of 62% and increase of 48% in throughput for this proposed topology compared to simple 2D mesh in cost of additional area overhead.
机译:随着芯片密度随着每一代工艺的增加而迅速增加,片上网络(NoC)的使用已成为SoC,MPSoC和大规模CMP(芯片多处理器)设计的普遍体系结构。研究人员已经提出了多种NoC解决方案,以满足增强的片上通信要求。在这里,底层网络互连体系结构(拓扑),路由器设计和路由策略对于提高整体系统性能起着重要作用。在这项工作中,已经提出了一种基于2D混合网格的星形拓扑,其目的是提供低延迟,低信道争用和更高吞吐量的基于系统的系统。观察到的实验结果表明,与简单的2D网格相比,此提议的拓扑结构最大的等待时间收益为62%,吞吐量增加了48%,这是额外的区域开销所致。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号