检索结果-南通市图书馆

HSRDN: High-Speed Router Design for Various No C Topologies

Chinese Journal of Electronics 2020年第2期29卷 281-290页

作者： E.Lakshmi Prasad M.N.Giri Prasad A.R.Reddy Department of ECE JNTUACE Department of RRC MITS

High-speed router design for network on chip(HSRDN) is proposed for controlling the traffic congestion and deadlocks. Diagonal based nearest-path routing algorithm for No C(DNRAN) can mitigate the effect of latency by opting for the nearest-path to reach the destination in a network and HSRDN is part of DNRAN. When we analyze the performance of DNRAN for all proposed topologies, nearly 50% better in terms of latency reduction and high throughput over existing router architectures. The proposed topologies(2 D-mesh,2 D-Star mesh over regional mesh(SMo RM), 3 D-mesh,and 3 D-torus) are tested with various applications, viz,audio, video and so on. Here, we also tested with cryptography application for DNRAN. When we analyzed the performance of experimental results, exclusively in 2 D-SMo RM nearly 0.6 times latency get reduced, area expanded by 0.25 and 0.33 times throughput increase in 2 D-SMo RM compared with 3 D-mesh and 3 D-torus.Therefore, DNRAN showed an exclusive performance in 2 D-SMo RM compared with other two topologies.

关键词： Logic Network on chip(NoC) virtual channels Latency Congestion Shortest path routing algorithm 3D-mesh 3D-torus

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Cell Rate Update Algorithm for ABR Traffic Control in ERICA Switch Scheme

Cell Rate Update Algorithm for ABR Traffic Control in ERICA ...

引用

1998 International Conference on Communication Technology

作者： Ilyoung Chong Sang Hong Lee Dept of Info.& Comms.Eng. Hankuk Univ.of Foreign Studies Korea Telecom

ISBN: (纸本)7800908275

＜正＞Source and destination behavior follows the description in the ATM Forum traffic management Specification version 4.0.The output buffer at each port is a FIFO queue shared by all VCs going through the port.Peerto -peer is simulated for persistent traffic and staggered traffic patterns.From simulation with different backbone link lengths,it appears that network scale does not have significant effect on the appropriate choice for averaging interval （AI） values as long as the number of VCs is the same.And a large AI gives better performance for a large number of VCs.Therefore,this paper proposes a novel dynamic AI algorithm, and its numerical results are presented.

关键词： asynchronous transfer mode packet switching telecommunication traffic telecommunication congestion control queueing theory telecommunication network management buffer storage ABR traffic control ERICA switch cell rate update algorithm destination behavior source behavior ATM Forum traffic management specification version 4.0 output buffer FIFO queue persistent traffic pattern staggered traffic pattern simulation backbone link lengths ATM network performance dynamic averaging integral algorithm numerical results virtual channels

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：