We use cookies to improve your experience with our site.
申小伟, 叶笑春, 谭旭, 王达, 张轮凯, 李文明, 张志敏, 范东睿, 孙凝晖. 一种面向数据流架构的高效片上路由结构[J]. 计算机科学技术学报, 2017, 32(1): 11-25. DOI: 10.1007/s11390-017-1703-5
引用本文: 申小伟, 叶笑春, 谭旭, 王达, 张轮凯, 李文明, 张志敏, 范东睿, 孙凝晖. 一种面向数据流架构的高效片上路由结构[J]. 计算机科学技术学报, 2017, 32(1): 11-25. DOI: 10.1007/s11390-017-1703-5
Xiao-Wei Shen, Xiao-Chun Ye, Xu Tan, Da Wang, Lunkai Zhang, Wen-Ming Li, Zhi-Min Zhang, Dong-Rui Fan, Ning-Hui Sun. An Efficient Network-on-Chip Router for Dataflow Architecture[J]. Journal of Computer Science and Technology, 2017, 32(1): 11-25. DOI: 10.1007/s11390-017-1703-5
Citation: Xiao-Wei Shen, Xiao-Chun Ye, Xu Tan, Da Wang, Lunkai Zhang, Wen-Ming Li, Zhi-Min Zhang, Dong-Rui Fan, Ning-Hui Sun. An Efficient Network-on-Chip Router for Dataflow Architecture[J]. Journal of Computer Science and Technology, 2017, 32(1): 11-25. DOI: 10.1007/s11390-017-1703-5

一种面向数据流架构的高效片上路由结构

An Efficient Network-on-Chip Router for Dataflow Architecture

  • 摘要: 在高性能计算领域中,数据流结构有较多的优势。在数据流计算模式中,大量的操作数需要通过片上网络在不同处理单元中传输。因此路由的设计对数据流结构的性能具有较大的影响。常见的路由是面向控制流多核处理器设计的,它们都不适合数据流结构。本文首先分析了数据流结构中片上网络的传输特征:多目的地,高注入率和性能对延迟敏感。基于这三项特征,本文提出了一种新型的面向数据流结构的片上网络路由结构,通过多目的地机制来实现单个数据多目的地的传输,通过输出缓冲机制提高路由的吞吐率,通过非分片机制来最小化路由传输时间。实验结果表明,相比现有面向控制流的路由结构,本文提出的路由结构将数据流结构的性能提高了3.6倍。

     

    Abstract: Dataflow architecture has shown its advantages in many high-performance computing cases. In dataflow computing, a large amount of data are frequently transferred among processing elements through the network-on-chip (NoC). Thus the router design has a significant impact on the performance of dataflow architecture. Common routers are designed for control-flow multi-core architecture and we find they are not suitable for dataflow architecture. In this work, we analyze and extract the features of data transfers in NoCs of dataflow architecture:multiple destinations, high injection rate, and performance sensitive to delay. Based on the three features, we propose a novel and efficient NoC router for dataflow architecture. The proposed router supports multi-destination; thus it can transfer data with multiple destinations in a single transfer. Moreover, the router adopts output buffer to maximize throughput and adopts non-flit packets to minimize transfer delay. Experimental results show that the proposed router can improve the performance of dataflow architecture by 3.6x over a state-of-the-art router.

     

/

返回文章
返回