期刊文献+

CICQ交换机中一类服务可保障的调度策略研究 被引量:3

A New Service Guaranteed Scheduling Policy for Buffered Crossbar Switches
下载PDF
导出
摘要 具备QoS保障能力的快速调度算法是高速交换机的首选.基于EPFTS(Ethernet-oriented physical frame timeslot switching)和CICQ(combined input-crosspoint-queued)交换技术的特点,提出了一类新的调度策略———TRWFS(ti meslot reservation weighted fair scheduling).为确保各端口对上保障业务的预留带宽,TRWFS以各端口对上保障业务预留时槽数为调度权重,以优先调度保障业务和平衡各保障业务的盈余时槽(surplus timeslot,定义为现实系统和理想系统之间的服务差额)为业务调度准则.基于该调度策略进一步提出了两种实现算法———TRWFS-I和TRWFS-II,总体上使实现TRWFS的时间复杂度降至O(1).性能分析和仿真实验结果均表明两种调度算法都达到了服务保障的设计目标,仿真实验结果还表明CICQ排队方式下与其他调度算法相比,TRWFS和轮询调度综合的调度机制具有交叉缓存容量要求更低的优点. Fast scheduler with service guarantee performance is preferred in high speed switching networks. Based on the features of EPFTS (Ethernet-oriented physical frame timeslot switching) and CICQ (combined input-crosspoint-queued), a new scheduling policy called timeslot reservation weighted fair scheduling (TRWFS) is introduced. In order to provide bandwidth guarantee service to guarantee- required (GR) traffic, TRWFS takes the total reserved timeslots of aggregated GR traffic on each I/O port pair as the basic scheduling weight and forwards traffic according to the following principles: 1) Giving dispatching priority to GR traffic and 2) Trying to balance the surplus timeslots (defined as service discrepancy between the actual packetized system and the idealized fluid system) for GR-traffic on each I/O port pair. In order to apply TRWFS and decrease the implementation complexity of TRWFS to constant time complexity O (1), two scheduling algorithms-TRWFS_ Ⅰ and TRWFS_Ⅱ are further presented. Analysis and simulation results demonstrate that the two TRWFS algorithms both reach the design objective. Moreover, simulation results show that TRWFS/round-robin combined scheduling scheme in CICQ switches require fewer crosspoint buffer compared with other schemes.
出处 《计算机研究与发展》 EI CSCD 北大核心 2007年第11期1873-1880,共8页 Journal of Computer Research and Development
基金 国家自然科学基金项目(60372065)
关键词 调度算法 服务质量 输入与交叉点综合的排队机制(CICQ) 面向以太网的物理帧时槽交换 (EPFTS) 盈余时槽 scheduling algorithm quality of service (QoS) combined input-crosspoint-queued (CICQ) Ethernet-oriented physical frame timeslot switching (EPFTS) surplus timeslot
  • 相关文献

参考文献14

  • 1H X Zeng,D Y Xu,J Dou.Promotion of physical frame timeslot switching (PFTS) over DWDM[J].Annual Review of Communications,2004,57:809-826
  • 2H X Zeng,J Dou,D Y Xu.Replace MPLS with EPFTS to build a SUPANET[C].In:Proc of the HPSR 2005.Piscataway:IEEE Press,2005.39-43
  • 3K Yoshigoe,K J Christensen.An evolution to crossbar switches with virtual output queuing and buffered cross points[J].IEEE Networks,2003,17(5):48-56
  • 4F Abel,C Minkenberg,et al.A four-terabit single-stage packet switch with large round-trip time support[C].Hot Interconnects'10,Symp on High Performance Interconnects (HOT-I'02),Palo Alto,CA,2002
  • 5R Rojas-Cessa,E Oki,H J Chao.On the combined inputcrosspoint buffered switch with round-robin arbitration[J].IEEE Trans on Communications,2005,53(11):1945-1951
  • 6T Javidi,R Magill,T Hrabik.A high-throughput scheduling algorithm for a buffered crossbar switches fabric[C].IEEE ICC,Helsinki,2001
  • 7M Nabeshima.Performance evaluation of a combined input and crosspoint-queued switch[J].IEICE Trans on Communications,2000,E83-B(3):737-741
  • 8M Katevenis,S Sidiropoulos,C Courcoubetis.Weighted roundrobin cell multiplexing in a general-purpose ATM switch chip[J].IEEE Journal Select Areas Communications,1991,9(8):1265-1279
  • 9A K Parekh,R G Gallager.A generalized processor sharing approach to flow control in integrated services networks:The single-node case[J].IEEE/ACM Trans on Networking,1993,1(3):344-357
  • 10Zhang Xiao,N B Laxmi.An efficient scheduling algorithm for combined input-crosspoint-queued (CICQ) switches[C].IEEE GLOBECOM,Dallas,Texas,2004

二级参考文献30

  • 1吴俊,陈晴,罗军舟.时隙间迭代的输入队列交换机Round-Robin调度算法[J].软件学报,2005,16(3):375-383. 被引量:11
  • 2T. E. Anderson, S. S. Owicki, J. B. Saxe, et al. High-speed switch scheduling for local area networks [J]. ACM Trans.Computer Systems, 1993, 11(4): 319--352
  • 3N, McKeown. The iSlip scheduling algorithm for input-queued switches[J]. IEEE/ACM Trans. Networking, 1999, 7(2): 188-201
  • 4R. Rojas-Cessa, E. Oki, Z. Jing, et al. CIXB-1: Combined input-one-cell-crosspoint buffered switch [C]. The Workshop on High Performance Switching and Routing, Dallas, USA, 2001
  • 5R. Rojas-Cessa, Round-robin selection with adaptable-size frame in a combined input-crosspoint buffered switch [J]. IEEE Communications Letters, 2003, 7( 11 ):555 - 557
  • 6T. Javidi, R. B. Magil, T. Hrabik. A high-throughput scheduling algorithm for a buffered crossbar switch fabric[C].The 2001 IEEE ICC, Dallas, USA, 2001
  • 7M. Nabeshima. Performance evaluation of a combined input-and crosspoint-queued switch [J]. IEICE Trans. Communications,2000, E83-B(3):737-741
  • 8L. Mhamdi, M. Hamdi. MCBF: A high-performance scheduling algorithm for buffered crossbar switches [ J ]. IEEE Communications Letters, 2003, 7(9): 431-433
  • 9Y. Tamir, G. L. Frazier. Dynamically-allocated multi-queue buffer for VLSI communication switches[J]. IEEE Trans.Computers, 1992, 41(6): 725-737
  • 10H. T. Kung, R. Morris. Credit-based flow control for ATM networks [J]. IEEE Network, 1995, 9(2) : 40-48

共引文献7

同被引文献31

  • 1曾华燊,窦军,汪海鹰.论“单物理层的用户数据传输平面体系结构网络”——SUPANET[J].计算机应用,2004,24(6):1-5. 被引量:8
  • 2吴俊,罗军舟.Randomized scheduling algorithm for input-queued switches[J].Journal of Southeast University(English Edition),2005,21(1):6-10. 被引量:1
  • 3许登元,张新有,刘文杰.物理帧时槽交换中改进的DWRR调度算法[J].西南交通大学学报,2005,40(6):735-739. 被引量:1
  • 4董雨果,汪胜荣,郭云飞,刘颖.带输入队列并行交换的负载平衡分析(英文)[J].软件学报,2007,18(2):229-235. 被引量:3
  • 5Zeng Huaxin,Xu Dengyuan,Dou Jun.On Physical Frame Time-slot Switching over DWDM[C]//PACAT03.IEEE press,Aug.2003:286-291.
  • 6Zeng Huaxin,Dou Jun,Xu Dengyuan.Single physical layer U-plane Architecture (SUPA)for Next Generation lnternet[R].Comprehensive Report on VoIP and enhanced IP Communications Services.IEC Publications,2004:197-227.
  • 7Guo Zirong,Zeng Huaxin.Simulation and analysis of weighted fair queueing algorithms in OPNET[C]//2009 International Conference on Computer Modeling and Simulation (ICCMS 2009).Macao:IEEE press,2009:114-118.
  • 8Dou Jun,Zeng Huaxin,Wang Haiying.Single User-Plane Architecture and its QoS Provisioning Mechanisms in Signaling and Management (S&M)Planes[C]//The 5th International Conference on Parallel and Distributed Computing,Applications and Technologies (PDCAT 04).Singapore,Dec.2004,429-440.
  • 9赵君 高雨.SUPANET中QoS协商流程研究.计算机科学,2004,31(8):21-24.
  • 10Zeng Huaxin,Gao Yu,Xia Yu.On NGN architecture and evolution strategy[C]//First ITU-T Kaleidoscope Academic Conference-Innovations in NGN:Future Network and Services.Geneva:IEEE press,May 2008:337-342.

引证文献3

二级引证文献5

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