A networks-on-chip （NoC） cost-effective design method was given based on the globallyasynchronous locally-synchronous （GALS） interconnect structure. In this method, the synchronous mode was used to transmit data among routers, network interface （NI）, and intellectual property （IP） via a synchronous circuit. Compared with traditional methods of implementing GALS, this method greatly reduces the transmission latency and is compatible with existing very large scale integration （VLSI） design tools. The platform designed based on the method can support two kinds of packetizing mechanisms, any topology, several kinds of traffic, and many configurable parameters such as the number of virtual channels, thus the platform is universal. An NoC evaluation methodology is given with a case study showing that the platform and evaluation methodology work well.
片上网络(network on chip,NoC)作为一种全新的片上互连通信架构,面积受限,却具有丰富的线资源。而且,三维片上网络的层间互连线很短,同时提供了在第三维度上的互连扩展性。根据这些特性,该文提出了一种基于三维Mesh片上网络的双链路互连架构。在垂直方向上,该架构采用双链路互连,使其通信带宽加倍;而且,跨层连接的垂直链路降低了消息传输的路由跳数。这些都带来网络平均延时的降低和最大吞吐量的提高,却仅仅增加一些控制逻辑电路。仿真结果验证了理论分析。与传统的单链路架构相比,该架构以较小的面积开销换取了较大的性能提高。
The network-on-chip （NoC） design methodology is an important trend for large system-on-chip designs to reduce the bandwidth and power constraints in traditional synchronous bus architectures. In the design of packet-based NoC, the packet-length plays an important role in the NoC throughput, latency, and energy consumption. The appropriate NoC packet-length was selected based on simulation and analysis of the packet-length effect on NoC for variable average data block length （ADBL） configuration parameters. A trade-off curve among throughput, latency, and energy consumption was developed and shows that the optimum packet length increases as the ADBL increases.
