共查询到19条相似文献,搜索用时 140 毫秒
1.
本文分析了一种具有点对多点服务能力的内部无阻塞输入/输出排队ATM交换机在反压控制下的性能指标。在每个输入端口信元的到达具有相同的强度,每个排头信元以一种相同的概率分布函数被复制成多个输往不同的输出端口的排头信元,复制后的排头柜元到达输出端口的概串相同为1/N,且输入、物出缓冲容量均为有限。为保证交换机内部不发生信元丢失,引入了反压机制(Backpressure)。本文利用矩阵几何分析法绘出了数值解,计算机仿真结果表明理论分析是正确的。 相似文献
2.
3.
输入/输出ATM交换机在突发性业务下的性能 总被引:1,自引:0,他引:1
本文详尽分析了内部无阻塞输入/输出排队反压型ATM交换机在突发性业务下信元丢失、交换机最大吞吐量等性能。输入端口信元的到达过程是ON-OFF突发流,且ON态以概率p发送信元,ON-OFF长度为Pareto分布的随机变量;属于同一突发流的信元输往同一个输出端口,不同突发流的信元等概率输往不同的输出端口;输入/输出缓冲器长度有限,交换机加速因子S任意。本文同时比较了突发长度为周期/几何分布下的交换机性能,其结论对实际设计一输入/输出排队反压型ATM交换机具有一定参考意义。 相似文献
4.
具有纵横输入互连方式和缓冲结构的递归Knockout交换网络 总被引:1,自引:0,他引:1
本文提出了具有纵横输入(CrosbarInput)互连方式和输入缓冲(InputBufered)结构的递归Knockout交换网络(CIBRKS).通过采用纵横输入互连方式可减少内部小交换单元的数目,并可使信元传送顺序不会受群输出端口数目的影响.而通过在每个输入端放置缓冲器可在保持丢失率性能不变的情况下,可使整个交换网络的级数减少,从而也就减少了信元在群网络中的传输时延.另外,在该结构中,通过把信元滤址的功能从每个小交换单元中提取出来放在每个输入端口,又进一步减少了小交换单元的功能.通过比较,我们认为,作为大规模ATM交换网络结构,CIBRKS结构比传统的RKS结构具有较好的性能/复杂度特性. 相似文献
5.
一种新的高性能ATM共享存储交换单元 总被引:1,自引:0,他引:1
ATM交换单元中.可选用三种基本排队策略一输入捧队、输出捧队和共享存贮捧队。根据平均信元时延、吞吐量和信元丢失概率性能.最佳的排队策略是共享存储捧队。由于传统的共享存储交换单元必须缓冲经过交换单元的所有信元,因此交换单元的吞吐量和信元丢失概率性能仍不理想,特别是对于大规模交换单元情况更是如此。本文提出了一种新的高性能共享存储交换单元,称为阻塞信元共享存储(BCSM)交换单元。顾名思义,BCSM仅仅缓冲交换单元输入端口的阻塞信元,而不必缓冲经过交换单元的所有信元。均匀业务下的分析结果证明,BCSM交换单元比一般共享存贮交换单元具有更好的性能。 相似文献
6.
本文利用矩阵几何分析法分析了内部无阻塞输入/输出排队反压型ATM交换机在均匀贝努利输入下的信元丢失、信元延时及吞吐量等性能指标。本文结论对实际设计一反压型输入/输出排队分组交换机具有一定参考意义。 相似文献
7.
本文利用矩阵几何分析法分析了内部无阻塞输入/输出排队反压型ATM交换机在均匀贝努利输出下的信元丢失、信元延时及吴吐量等性能指标。本文结论对实际设计一反压型输入/输出排队分组交换机具有一定参考意义。 相似文献
8.
本文分析了一种输入排队缓冲器有丢失优先级的内部无阻塞输入/输出排队ATM交换机在反压控制下的信元丢失指标。在每个输入端口高、低优先级信元的到达具有相同的强度,到达输出端口的概率相同为1N,且输入、输出缓冲容量均为有限。为保证交换机内部不发生信元丢失,引入了反压机制(Backpressure)。文中表明,使用丢失优先级策略的交换机比纯输入/输出排队交换机更能满足不同业务服务等级QoS(QualityofService)的丢失要求,而且所需缓冲容量减少。 相似文献
9.
10.
11.
The purpose of this paper is to develop a global design for test methodology for testing a core-based system in its entirety. This is achieved by introducing a bypass mode for each core by which the data can be transferred from a core input port to the output port without interfering the core circuitry itself. The interconnections are thoroughly tested because they are used to propagate test data (patterns or signatures) in the system. The system is modeled as a directed weighted graph in which the accessibility (of the core input and output ports) is solved as a shortest path problem. Finally, a pipelined test schedule is made to overlap accessing input ports (to send test patterns) and output ports (to observe the signatures). The experimental results show higher fault coverage and shorter test time. 相似文献
12.
13.
By using silicon-on-insulator(SOI) platform, 12 channel waveguides, and four parallel-coupling one-microring resonator routing elements, a non-blocking four-port optical router is proposed. Structure design and optimization are performed on the routing elements at 1 550 nm. At drop state with a power consumption of 0 m W, the insertion loss of the drop port is less than 1.12 d B, and the crosstalk between the two output ports is less than-28 d B; at through state with a power consumption of 22 m W, the insertion loss of the through port is less than 0.45 d B, and the crosstalk between the two output ports is below-21 d B. Routing topology and function are demonstrated for the four-port optical router. The router can work at nine non-blocking routing states using the thermo-optic(TO) effect of silicon for tuning the resonance of each switching element. Detailed characterizations are presented, including output spectrum, insertion loss, and crosstalk. According to the analysis on all the data links of the router, the insertion loss is within the range of 0.13—3.36 d B, and the crosstalk is less than-19.46 d B. The router can meet the need of large-scale optical network-on-chip(ONo C). 相似文献
14.
The bufferless router emerges as an interesting option for cost-efficient in network-on-chip (NoC) design. However, the bufferless router only works well under low network load because deflection more easily occurs as the injection rate increases. In this paper, we propose a load balancing bufferless deflection router (LBBDR) for NoC that relieves the effect of deflection in bufferless NoC. The proposed LBBDR employs a balance toggle identifier in the source router to control the initial routing direction of X or Y for a flit in the network. Based on this mechanism, the flit is routed according to XY or YX routing in the network afterward. When two or more flits contend the same one desired output port a priority policy called nearer-first is used to address output ports allocation contention. Simulation results show that the proposed LBBDR yields an improvement of routing performance over the reported bufferless routing in the flit deflection rate, average packet latency and throughput by up to 13%, 10% and 6% respectively. The layout area and power consumption compared with the reported schemes are 12% and 7% less respectively. 相似文献
15.
16.
17.
C.R. Doerr M. Zirngibl C.H. Joyner 《Photonics Technology Letters, IEEE》1996,8(4):500-502
Because the transmission through each input-output combination of a conventional waveguide grating router consists of a periodic sequence of equal-height passbands spaced by the router free spectral range, the absolute laser oscillation wavelength and single-passband oscillation can be difficult to control for all the channels in the multifrequency waveguide grating router laser. However, the router can be designed to have one dominant passband, with the neighboring passbands experiencing lower transmission coefficients for all the input output combinations by means of chirping of the waveguide grating. Such chirping is discussed theoretically and is demonstrated experimentally. 相似文献
18.
19.
We consider cell-based switch and router architectures whose internal switching matrix does not provide enough speed to avoid input buffering. These architectures require a scheduling algorithm to select at each slot a subset of input buffered cells which can be transferred toward output ports. We propose several classes of scheduling algorithms whose stability properties are studied using analytical techniques mainly based upon Lyapunov functions. Original stability conditions are also derived for scheduling algorithms that are being used today in high-performance switch and router architectures 相似文献