期刊文献+

一种维序的基于组合输入输出排队的并行交换结构 被引量:7

A Parallel Packet Switch Achieving In-Order Cell Delivery with Combined-Input-and-Output Queuing Switches
下载PDF
导出
摘要 提出一种按序排队(in-order queuing,简称IOQ)PPS体系结构,通过在分流控制器引入固定尺寸的缓冲区,实现负载在每个交换平面的均匀分配;中间层组合输入输出排队(combined input-and-output queuing,简称CIOQ)交换平面受控于中央调度器,在每个时间槽(timeslot),中央调度器将同一种匹配实施到每一个交换平面,称之为同步调度策略.可以证明,在该体系结构下,轮询(round robin)分派算法配合同步调度策略可以保证同一条流的信元按序从交换平面读出.进一步提出了严格最长队列优先同步调度算法,极大地减少了中央调度器需要维护的状态信息和信元重定序开销.与目前主流的PPS设计相比,IOQPPS(in-order queuing parallel packet switch)实现机制简单,易于硬件实现.模拟结果表明,IOQPPS具有最优的延迟性能. An in-order queuing (IOQ) PPS architecture proposed in this paper uses a small fixed-size buffer in the demultiplexor to distribute traffic equally among switch planes, with central combined input-and-output queuing (CIOQ) switch planes under the control of a single scheduler that applies the same matching at each of the parallel switch planes during each cell slot. This operation is called synchronous scheduling. It is proved that the round robin demultiplexing algorithm along with synchronous scheduling guarantees cells of a flow can be read in order from the output queues of the switch planes. Furthermore, by using a synchronous scheduling called strict longest queue first (SLQF) algorithm this scheme reduces considerably not only the amount of state information required by the scheduler, but the communication overhead required to achieve cell reordering. Compared with existing PPS designs, IOQ PPS (in-order queuing parallel packet switch) is more practical to implement in hardware because of its simple implementation mechanisms, as the experimental results demonstrate, and it offers the best delay performance.
出处 《软件学报》 EI CSCD 北大核心 2008年第12期3207-3217,共11页 Journal of Software
基金 国家自然科学基金No.90604006 国家重点基础研究发展计划(973)No.2003CB314802~~
关键词 交换结构 IOQ PPS(in-order QUEUING PARALLEL PACKET switch) 组合输入输出排队交叉开关 并行报文交换 信元按序发送 switch architecture IOQ PPS (in-order queuing parallel packet switch) CIOQ switch parallel packet switch in-order cell delivery
  • 相关文献

参考文献20

  • 1McKeown N. Memory for High Performance Internet Routers, Presentation to Micron. 2003. http://tiny-tera.stanford.edu/-nickm/ talks/Micron Feb 2003.ppt.
  • 2Iyer S, McKeown NW. Analysis of the parallel packet switch architecture. IEEE/ACM Trans. on Networking, 2003,11(2):314-324.
  • 3Aslam A, Christensen K J. A parallel packet switch with multiplexors containing virtual input queues. Computer Communications, 2004,27:1248-1263.
  • 4Cuppu V, Jacob B, Davis B. A performance comparison of contemporary DRAM architectures. In: Proc. of the 26th Int'l Symp. on Computer Architecture (ISCA'99). 1999. 222-233. http://www.ece.umd.edu/-blj/papers/isca99.pdf.
  • 5http://www.rambus.com.
  • 6Iyer S, Kompella R, McKeown N. Analysis of a memory architecture for fast packet buffers. In: Proc. of the IEEE Workshop on High Performance Switching and Routing. 2001. 368-373. http://klamath.stanford.edu/-nickm/papers/mmahpsr01.pdf.
  • 7Bianco A, Giaccone P, Leonardi E, Neff F. A framework for differential frame-based matching algorithms in input-queued switches. In: Proc. of the IEEE INFOCOM. Piscataway: Institute of Electrical and Electronics Engineers Inc., 2004. 1147-1157.
  • 8Oki E, Rojas-Cessa R, Chao H J. A pipeline-based maximal-sized matching scheme for high-speed Input-buffered switches. IEICE Trans. on Communications, 2002,E85-B(7):1302-1311.
  • 9Chang CS, Lee DS, Jou YS. Load balanced Birkhoff-von Neumann switches part I: One-Stage buffering. Computer Communications, 2002,25(6):611-622.
  • 10Chiussi F, Khotimsky D, Krishnan S. Generalized inverse multiplexing of switched ATM connections. In: Proc. of the IEEE GLOBECOM. Piscataway: Institute of Electrical and Electronics Engineers Inc., 1998.3134-3140.

同被引文献62

  • 1李文杰,刘斌,徐扬,廖恒.Parallel Switch System with QoS Guarantee for Real-Time Traffic[J].Journal of Computer Science & Technology,2006,21(6):1012-1021. 被引量:1
  • 2Yan B. Supply Chain Information Transmission based on RFID and Intemet of Things. 2009.
  • 3Michael M P. Architectural Solutions for Mobile RFID Services for the Internet of Things, 2008.
  • 4Steens W R著.范建华等译.TCP/IP详解卷1:协议.北京:机械工业出版社,1999.
  • 5Comer D E著.林瑶等译.用TCP/IP进行网际互联第一卷:原理、协议与结构(第四版).北京:电子工业出版社,2004.
  • 6MCKEOWN N,IZZARD M,MEKKITTIKUL A,et al.The tiny tera:a packet switch core[J].IEEE Micro,1997,17(1):26-33.
  • 7Digital Equipment Corporation.G1GAswitch[EB/OL].http://www.networks.digital.com,1997.
  • 8Ascend Communications.GRF family of switches[EB/OL] ,http://www.ascend.com,1998.
  • 9MCKEOWN N.The iSLIP scheduling algorithm for input-queued switches[J].IEEE/ACM Transactions on Networking,1999,7(2):188-201.
  • 10SERPANOS D N,ANTONIADIS P I.FIRM:a class of distributed scheduling algorithms for high-speed ATM switches with multiple input queues[A].Proceedings of IEEE INFOCOM'00[C].Tel Aviv,Israel,2000.548-554.

引证文献7

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部