您现在的位置: 首页> 研究主题> 交叉开关

交叉开关

交叉开关的相关文献在1994年到2022年内共计139篇,主要集中在自动化技术、计算机技术、无线电电子学、电信技术、力学 等领域,其中期刊论文73篇、会议论文10篇、专利文献265963篇;相关期刊54种,包括探测与控制学报、今日电子、数字技术与应用等; 相关会议10种,包括全国抗恶劣环境计算机第二十五届学术年会 、第十九届计算机工程与工艺年会暨第五届微处理器技术论坛、第十六届计算机工程与工艺年会暨第二届微处理器技术论坛等;交叉开关的相关文献由295位作者贡献,包括张雨生、田佩佳、葛宁等。

交叉开关—发文量

期刊论文>

论文:73 占比:0.03%

会议论文>

论文:10 占比:0.00%

专利文献>

论文:265963 占比:99.97%

总计:266046篇

交叉开关—发文趋势图

交叉开关

-研究学者

  • 张雨生
  • 田佩佳
  • 葛宁
  • 蔡凯
  • 冯丹
  • 刘景宁
  • 张扬
  • 童薇
  • 梁华国
  • 欧阳一鸣
  • 期刊论文
  • 会议论文
  • 专利文献

搜索

排序:

年份

    • SUN Xue; CAO Suzhi; XU Hui
    • 摘要: 为了实现光纤通道交换机中不同长度多优先级信元的联合调度,提出并实现了一种支持变长信元的基于CROSSBAR交换结构的多优先级输入排队调度策略,以支持关键数据能以最快的速度被调度转发.通过采用虚拟输出队列(VOQ)机制、CROSSBAR交换结构以及流水线设计来提高光纤通道交换机的性能.使用FPGA对光纤通道交换机实现了文章所提出的调度策略并进行了测试,实现的光纤通道交换机最多支持6个不同优先级,支持信元长度为16~2148字节的联合调度.
    • 欧阳一鸣; 笪健; 李建华; 梁华国; 黄正峰; 杜高明
    • 摘要: Crossbar is an important part of the router in network-on-chip, when it breaks down, the data can't be transmitted to the next hop which will cause waste of resources and influence the performance of entire network. Considering faults of crossbar internal component including Demultiplexers, internal links and Multiplexers, in order to make full use of available resources of crossbar, this article proposes a fine-grained crossbar fault-tolerant design which can achieve the diversity of fault-tolerant path. Firstly, the data arrive to Demultiplexer, seeing the fault information table, and then choosing a trouble-free Demultiplexer orderly for data transmitting when there is fault. Secondly, the data arrive to internal links through Demultiplexer, choosing a suitable fault-tolerant link from the fault information table in the same way. Finally, the data ar-rive to Multiplexer. When faults occur, the data can be transmitted to the output port by the extra Multiplexer. Experimental results show that under the condition of the failure rate increase gradually, compared to com-parative methods, our scheme can reduce average network latency by 6%~36%, increase throughput rate by 6.7%~27.7%, and the fault-tolerant effect will not decrease significantly with increasing failure rate.%在片上网络中,交叉开关是路由器的重要组成部分,当其出现故障会导致数据无法传到下一跳,易造成资源浪费,影响整个网络的性能.对此,通过详细考虑交叉开关中数据分配器、内部链路以及数据选择器的故障情况,充分利用交叉开关的可用资源,提出交叉开关细粒度容错设计方案,实现容错路径的多样性.当数据到达数据分配器时,查看故障信息表,若存在故障则顺序选择一个无故障的数据分配器传输数据;当数据经过数据分配器传至内部链路时,若存在链路故障,同样查找故障信息表快速选择合适的容错链路进行传输;最后数据到达数据选择器,若此时数据选择器出现故障,数据则经过额外的数据选择器到达输出端口.实验结果表明,故障率逐渐增大的时候,该方案比已有方法平均延时降低了6%~36%,吞吐率提高了6.7%~27.7%,且容错效果不会随着故障率增大而显著下降.
    • 李玉发; 王红春
    • 摘要: 基于VOQ的 crossbar交换结构在高速互联网络中应用十分广泛。现有的VOQ调度算法以迭代方式工作,通常分为三步:请求、授权和接受。通过将仲裁合并到请求步骤,可以省略接受步骤,用来实现两步迭代匹配算法。提出了一种适用于两步迭代匹配算法的高效硬件实现结构,并进行了大量的仿真。仿真结果显示,两步迭代匹配算法可以达到与三步迭代匹配算法类似的性能。%Virtual output queued ( VOQ) crossbar switches have been widely used as high speed intercon-nects . Existing practical VOQ scheduling algorithms work in an iterative manner and each iteration usually includes three steps:request,grant and accept. By incorporating arbitration into the request step the ac-cept step can be eliminated,and two-step iterative matching can be achieved. We also propose a hard-ware efficient implementation for the two-step iterative matching algorithms,and conduct extensive simu-lations. The simulation results demonstrate that two-step algorithms have almost the same performance as that of three-step algorithms.
    • 刘怡俊; 刘俊辉
    • 摘要: 针对现有拓扑网络路由节点传输数据时跳数过多,网络直径大和负载不均匀等问题,采用一种多点专用通道( MECS)拓扑网络结构,在XY方向采用广播方式,使源地址到目的地址最多经过2个路由器节点。在原基础上对输入缓冲区用虚拟通道改进,输出端用交叉开关调度控制并行执行,同时对需要实时通信的业务进行优先传输,达到了缩短网络直径,降低功耗的目的。用Modelsim仿真可知,实现了减少平均时延,提高吞吐量,而且提高了网络性能和服务质量的目的。%In order to solve the problems of too many hops in the routing node of the existing network to-pology, the longer network diameter and the imbalance load, a new MECS Topology Network, which broadcasts in the XY direction, is used.So the data from source address to the target address are only through a maximum of two router nodes.We adopt the form of input buffer with a virtual channel based on the original and the output buffer under the crossbar scheduling control so they can execute in a paral-lel way, while the business can prioritize the transmission of real-time communication and reduce power consumption.The Modelsim simulation shows that the Routing Node can achieve a shorter average delay, increase throughput, improve the network performance and quality of service.
    • 欧阳一鸣; 何敏; 梁华国; 汪秀敏; 常郝
    • 摘要: 针对因路由器内部输入缓存和交叉开关故障引起的可靠性及网络拥塞问题,提出一种故障感知的RVOQ容错架构设计方案.首先在输入端口处增加冗余虚通道进行输入缓存故障的容错设计,通过故障信息的反馈和仲裁算法使得数据选择有效的路径进行传输;然后修改交叉开关的架构,增加多路选择开关和相应控制模块,输入数据优先考虑本地数据链路,故障情况下选择冗余路径进行数据传输.实验结果表明,在故障数为3时,该方案比已有方法的时延降低了11%~53.1%;在网络出现多个故障、面临网络重负载时,仍然能够保证系统的高可靠性以及传输性能.%Aiming at the reliability and congestion problems caused by the faults occurring in input buffer and crossbar of the router, this article proposes a fault-tolerant and fault-aware RVOQ router design to toler-ate fault. Firstly, this article adds a redundant virtual channel in each input port to tolerate input buffer faults, through the feedback of fault information and arbitration algorithm, our fault-tolerant architecture design can ensure data select valid path to transmit. Secondly, this article modifies the crossbar architecture, by adding a multi-channel selection switch and corresponding control modules, input data gives priority to the local data channel, when faults occur, the data select redundant path to transmit. The experimental results show that, compared to the reference documentation, our proposed scheme has 11%−53.1% less average network la-tency in the presence of 3 faults under uniform traffic pattern. Our scheme has obvious advantage and en-sures the high-reliability and low-latency of the entire network when congestion and faults occur.
    • 张铭泉; 古志民; 赵鑫; 张吉赞
    • 摘要: 为减少嵌入式系统片上多核互连网络的动态能耗,提出了一种基于频繁交换值的多核交叉开关节能方法.利用片上多核互连网络中值的局部性,设计了频繁交换值缓存(FEVC),通过减少互连链路上的通信量和位变换数,有效降低了片上交叉开关互连网络的动态能耗.为达到最佳节能效果,通过实验确定了FEVC中保存的值的个数.实验结果表明与原系统相比,在单独使用频繁交换值缓存,只保存4个数据值时可以实现节能13%,结合翻转码算法可使节能比例达到20%.
    • 周婷; 赵有健; 王瑞生
    • 摘要: The Internet growth coupled with the vaxiety of multicast services is creating an increasing need for multicast traffic support by routers and packet switches. However, the traditional crossbar-based multicast scheduling schemes are unable to meet the needs of high-speed switching for the low performance and high implementation complexity. In this paper, a Two-Stage Multicast Switch (TSMS) is proposed, which is a serial combination of a Multicast To Unicast (MTU) switch to copy input cells from various sources simultaneously and a Combined Input and Output Queueing (CIOQ) switch to deliver copies of multicast cells to their final destinations. Based on MTU switch, a novel Largest Fanout Cardinal Number First-Uniform Middle Buffer Allocation (LFCNF-UMBA) scheduling algorithm is designed to determine how to copy multicast cells into unicast cells. By coordinately using Maximal Matching scheduling algorithm in CIOQ switch, it is proved that speedup of 2 -2/(N+1) is necessary and 2 is sufficient for a M×N TSMS to achieve 100% throughput under any admissible multicast traffic pattern, which is also verified by the simulation results.%在路由器或交换机的交换结构中实现组播是提高组播应用速度的重要途径之一.传统的交叉开关结构(crossbar)组播调度方案有两种缺陷,一种是性能较低,另一种是实现的复杂度太高,无法满足高速交换的需要.该文提出了一个新的基于交叉开关的两级组播交换结构(TSMS),第1级是组播到单播的交换结构,第2级是联合输入和输出排队(CIOQ)交换,并为该结构设计了合适的最大扇出排队(FCN)优先-均匀分配中间缓存调度算法(LFCNF-UMBA).理论分析和仿真实验都显示在该结构中,加速比低于2- 2/(N+1)倍时吞吐率不可能实现100%;而采用LFCNF-UMBA调度算法,2倍加速比就可保证在任意允许(admissible)组播的吞吐率达到100%.
    • 刘涌; 徐厚生; 王兴伟
    • 摘要: The roaring growth of multimedia data demands cost-effective storage solutions in urgent, so that a near-line storage system based on crossbar switch was presented. High-capacity SATA disk array was used in this system instead of tape library, and connects on-demand with the host was implemented through the crossbar switch instead of manipulator. The storage system model was proposed. The hardware and firmware design of crossbar switch matrix, scheduling policy of the host, address-planning of system port, hardware port and internal switch were respectively introduced. The prototype system showed that the proposed system could meet the demand for massive multimedia data storage.%针对多媒体数据的爆炸式增长急需高性价比的存贮解决方案,提出了一种基于交叉开关的近线存贮系统.该系统利用大容量的SATA硬盘阵列取代机械手控制的磁带库,通过交叉开关矩阵切换实现与主机按需连接.给出了存贮系统的模型,对交叉开关矩阵的硬件和固件设计、主机调度策略的设计思想、系统端口编址、硬盘端口编址、内部开关编址分别进行了阐述.原型系统验证本系统可以很好地满足多媒体业务的需求.
    • 汪健; 张磊; 赵忠惠; 王少轩; 陈亚宁
    • 摘要: In recent years, Multi-core architectures have been the mainstream design of the processors as well as the mainstream processing platform of various communication and multimedia applications. The communication structure becomes one of the core technologies of Multi-core processors. The efficiency of communication between the inter-core is the key indicator of Multi-core processor performance. There are three main types of communication ar- chitecture : Bus system, Cross-bar network and On-Chip Network. Bus structure designs have the advantages of con- venience, low hardware consumption and low cost. Cross-bar design is suitable for the switching network structure of building large-capacity systems. On-Chip Network is larger and more sophysicated on-chip network, which can com- pletely solve the problem of Multi-core architecture and therefore becomes one of the most promising solutions for the Multi-core system. The paper discusses the principle, system structure and function of the NOC communication ar- chitecture and presents the realization of some units.%多核处理器已经成为处理器的主流,并发展成为各种通信与媒体应用的主流处理平台。通讯结构是多核系统中的核心技术之一,核问通信的效率是影响多核处理器性能的重要指标。目前有3种主要的通讯架构:总线系统结构、交叉开关网络和片上网络。总线结构设计相对方便、硬件消耗较少、成本较低;交叉开关是适合用于构建大容量系统的交换网络结构;而片上网络是更高层次、更大规模的片上网络系统,目前可以解决多核体系结构问题,是多核系统最有前途的解决方案之一。文中在分析了NoC结构的基本原理、系统结构和功能的同时,也提供了部分单元的设计实现。
  • 查看更多

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