Design and performance evaluation of unbuffered self-routingnetworks for wide-band packet switching

The design of several unbuffered self-routing networks for wideband packet switching is presented. These networks possess many attractive characteristics, including simple node design, simple fault diagnosis, tolerance of faults, tolerance of unbalanced load, and packets delivered in sequence. It is shown, through analysis and simulation, that the maximum throughput of an unbuffered self-routing network is greater for fixed packet lengths than for exponentially distributed packet lengths. This means that the maximum throughput of an unbuffered self-routing network will be different for such applications as ATM (asynchronous transfer mode) and frame-relay, if they do not have the same packet length distribution     

New performance bounds of a class off self-routing networks

Conventional bounds for the maximum throughput of an unbuffered banyan network and its topology equivalents are based on the parallel path setup (PPS) assumption, i.e. paths are set up simultaneously. But the bounds derived under the PPS assumption can be surpassed with a slight variation on the path setup scheme. The authors study the maximum throughputs of an unbuffered banyan network and its topology equivalents in an incremental path setup (IPS) environment. The results represent the ultimate bounds for the maximum throughputs of unbuffered banyan networks. Although the exact analysis of this problem involves a combinatorial explosion, an approximate analysis is given in this work to verify the simulation results     

Multistage network with globally controlled switching stages and its implementation using optical multi-interconnection modules

Plane-to-plane guided-wave-based interconnection modules are proposed as building blocks for scalable optoelectronic multistage interconnection networks (MINs). This approach leads naturally to a MIN paradigm based not on cascading switching stages containing several size-reduced crossbars, as in the shuffle-exchange (SE) networks, but on cascading permutation-reduced crossbars instead, one per stage. The interest of such an architecture lies in the control simplicity and scalability potential. Transparent circuit switching for permutation routing is possible in such an unbuffered "globally switched" multistage interconnection network (GSMIN). Preliminary experiments using fiber-based interconnection modules are presented. Performance analysis and simulation of a buffered GSMIN is also studied for packet routing purposes.     

Queueing analysis of scheduling policies in copy networks ofspace-based multicast packet switches

Space-based multicast switches use copy networks to generate the copies requested by the input packets. In this paper our interest is in the multicast switch proposed by Lee (1988). The order in which the copy requests of the input ports are served is determined by the copy scheduling policy and this plays a major part in defining the performance characteristics of a multicast switch. In any slot, the sum of the number of copies requested by the active inputs of the copy network may exceed the number of output ports and some of the copy requests may need to be dropped or buffered. We first propose an exact model to calculate the overflow probabilities in an unbuffered Lee's copy network. Our exact results improve upon the Chernoff bounds on the overflow probability given by Lee by a factor of more than 10. Next, we consider buffered inputs and propose queueing models for the copy network for three scheduling policies: cyclic service of the input ports with and without fanout splitting of copy requests and acyclic service without fanout splitting. These queueing models obtain the average delay experienced by the copy requests. We also obtain the sustainable throughput of a copy network, the maximum load that can be applied to all the input ports without causing an unstable queue at any of the inputs, for the scheduling policies mentioned above     

An approximate analysis of the performance of deflection routing inregular networks

Regular two-dimensional architectures are being considered as alternatives to the linear topology metropolitan area networks (MANs) that are popular today. Deflection routing is an adaptive routing strategy that performs well on such architectures. A general analytic model has been developed to study the performance of buffered deflection routing in regular networks. The Manhattan street network, the ShuffleNet, and the shuffle exchange network have been studied as candidate two-connected networks with different topological characteristics. The results show that deflection routing performs well on both the Manhattan street network and the ShuffleNet, even under heavy loads, while on the shuffle exchange network it does not perform as well. The introduction of just a few buffers provides significant improvement in the delay-throughput performance over unbuffered deflection routing, especially in networks with large propagation delays. The analytic results are found to match the simulations very closely in most cases     

A new approach for allocating buffers and bandwidth toheterogeneous, regulated traffic in an ATM node

A new approach to determining the admissibility of variable bit rate (VBR) traffic in buffered digital networks is developed. In this approach all traffic presented to the network is assumed to have been subjected to leaky-bucket regulation, and extremal, periodic, on-off regulated traffic is considered; the analysis is based on fluid models. Each regulated traffic stream is allocated bandwidth and buffer resources which are independent of other traffic. Bandwidth and buffer allocations are traded off in a manner optimal for an adversarial situation involving minimal knowledge of other traffic. This leads to a single-resource statistical-multiplexing problem which is solved using techniques previously used for unbuffered traffic. VBR traffic is found to be divisible into two classes, one for which statistical multiplexing is effective and one for which statistical multiplexing is ineffective in the sense that accepting small losses provides no advantage over lossless performance. The boundary of the set of admissible traffic sources is examined, and is found to be sufficiently linear that an effective bandwidth can be meaningfully assigned to each VBR source, so long as only statistically-multiplexable sources are considered, or only nonstatistically-multiplexable sources are considered. If these two types of sources are intermixed, then nonlinear interactions occur and fewer sources can be admitted than a linear theory would predict. A qualitative characterization of the nonlinearities is presented. The complete analysis involves conservative approximations; however, admission decisions based on this work are expected to be less overly conservative than decisions based on alternative approaches     

Performance of buffered banyan networks under nonuniform trafficpatterns

An analytical method of evaluating the performance of the buffered banyan packet-switching network under nonuniform traffic patterns is presented. It is shown that nonuniform traffic can have a detrimental effect on the performance of the network. The analytical model is extended to evaluate the performance of multibuffer and parallel banyan networks. These modified networks are shown to have better throughput capacity and delay performance than the single-buffer banyan network     

An inversion algorithm to compute blocking probabilities in lossnetworks with state-dependent rates

The algorithm developed in Choudhury et al. (1994) for computing (exact) steady-state blocking probabilities for each class in product-form loss networks is extended to cover general state-dependent arrival and service rates. This generalization allows to consider, for the first time, a wide variety of buffered and unbuffered resource-sharing models with non-Poisson traffic, as may arise with overflows in the context of alternative routing. As before, the authors consider noncomplete-sharing policies involving upper-limit and guaranteed-minimum bounds for the different classes, but in the present paper both bounds are discussed simultaneously. These bounds are important for providing different grades of service with protection against overloads by other classes. The algorithm is based on numerically inverting the generating function of the normalization constant, which is derived in the present paper. Major features of the algorithm are: dimension reduction by elimination of nonbinding resources and by conditional decomposition based on special structure, an effective scaling algorithm to control errors in the inversion, efficient treatment of multiple classes with identical parameters and truncation of large sums. The authors show that the computational complexity of the inversion approach is usually significantly lower than the alternative recursive approach     

Traffic grooming,routing, and wavelength assignment in an optical WDM mesh networks based on clique partitioning

In wavelength routed optical networks, the number of wavelength channels is limited due to several constraints and each wavelength as well as each lightpath support traffic in the Gbps range. On the other hand, the traffic requested by an individual connection is still in the Mbps range. Therefore, to utilize the network resources (such as bandwidth and transceivers) effectively, several low-speed traffic streams have to be efficiently groomed or multiplexed into one or more high-speed lightpaths. The grooming problem of a static demand is considered as an optimization problem. In this work, we have investigated the traffic grooming problem with the objective of maximizing the network throughput for wavelength-routed mesh networks and map this problem to the clique partitioning problem. We have proposed an algorithm to handle general multi-hop static traffic grooming based on the clique partitioning concept. The efficiency of our approach has been established through extensive simulation on different sets of traffic demands with different bandwidth granularities for different network topologies and compared the approach with existing algorithms.     

Mobility-based CAC algorithm for arbitrary call-arrival rates in CDMA cellular systems

This paper presents a novel approach for designing a call-admission control (CAC) algorithm for code-division multiple-access (CDMA) networks with arbitrary call-arrival rates. The design of the CAC algorithm uses global information; it incorporates the call-arrival rates and the user mobilities across the network and guarantees the users' quality of service (QoS) as well as prespecified blocking probabilities. On the other hand, its implementation in each cell uses local information; it only requires the number of calls currently active in that cell. We present several cases for a nontrivial network topology where our CAC algorithm guarantees QoS and blocking probabilities while achieving significantly higher throughput than that achieved by traditional techniques. We also calculate the network capacity, i.e., the maximum throughput for the entire network, for prespecified blocking probabilities and QoS requirements.     

A Network Coding Scheme to Improve Throughput for IEEE 802.11 WLAN

In IEEE 802.11 infrastructure wireless local area network (WLAN), the communication between any two nodes is relayed by an access point (AP), which becomes the bottleneck of WLAN and severely restricts the overall throughput. It is well known that network coding technique is able to greatly improve the throughput of wireless networks. But, the available coding schemes do not make full advantage of channel capacity due to the fact that they pick at most one packet from each data flow for coding and the picked packets may have a great difference in packet size, wasting some channel capacity. To remedy the problem, in this paper, we propose the coding scheme that combines multiple buffered packets in one flow into a larger packet for coding so that the packets participating in coding have close sizes. We formulate an integer programming problem to find the optimal packet coding, which is solved by an optimal algorithm with relative high time complexity together with a heuristic algorithm with relative low time complexity. Simulation results show that the proposed coding scheme is able to greatly improve the throughput of WLAN and the throughput gain increases with the growth of the number of coding flows.     

A medium access protocol exploiting multiuser-detection in CDMA ad-hoc networks

A new medium access protocol which exploits the physical layer capability of multiuser detection is proposed to help in improving the throughput/delay performance of ad-hoc networks. When more than one node has packets buffered for a common node in the neighborhood, all such nodes can simultaneously transmit their packets to the common receiver after reserving their surrounding channel. This is achieved in our protocol by extending the (sender-initiated) CSMA/CA collision avoidance framework by the receiver-initiated medium access technique and incorporating the transmission power control. We analyze the improvement in the throughput that can be achieved over the basic sender-initiated collision avoidance protocol in the network. Since the throughput improvement via multi-packet reception is influenced by the network layer activity as well, the performance of our protocol rolls back to that of the basic sender-initiated protocol in case of no coordination from the network layer. For the evaluation of performance of our protocol we simulate ad-hoc networks for different network topologies and traffic configurations. We observe the scheme to be capable in significantly improving the throughput/delay performance of the network.     

Omega network-based ATM switch with neural network-controlledbypass queueing and multiplexing

Multistage interconnection networks (MINs) have long been studied for use in switching networks. Since they have a unique path between source and destination and the intermediate nodes of the paths are shared, internal blocking can cause very poor throughput. This paper proposes a high throughput ATM switch consisting of an Omega network with a new form of input queues called bypass queues. We also improve the switch throughput by partitioning the Input buffers into disjoint buffer sets and multiplexing several sets of nonblocking cells within a time slot, assuming that the routing switch operates only a couple of times faster than the transmission rate. A neural network model is presented as a controller for cell scheduling and multiplexing in the switch. Our simulation results under uniform traffic show that the proposed approach achieves almost 100% of potential switch throughput     

The AWG/spl par/PSC network:a performance-enhanced single-hop WDM network with heterogeneous protection

Single-hop wavelength-division-multiplexing (WDM) networks based on a central passive star coupler (PSC) or arrayed-waveguide grating (AWG) hub have received a great deal of attention as promising solutions for the quickly increasing traffic in metropolitan and local area networks. These single-hop networks suffer from a single point of failure: if the central hub fails, then all network connectivity is lost. To address this single point of failure in an efficient manner, we propose a novel single-hop WDM network, the AWG/spl par/PSC network. The AWG/spl par/PSC network consists of an AWG in parallel with a PSC. The AWG and PSC provide heterogeneous protection for each other; the AWG/spl par/PSC network remains functional when either the AWG or the PSC fails. If both AWG and PSC are functional, the AWG/spl par/PSC network uniquely combines the respective strengths of the two devices. By means of analysis and verifying simulations we find that the throughput of the AWG/spl par/PSC network is significantly larger than the total throughput obtained by combining the throughput of a stand-alone AWG network with the throughput of a stand-alone PSC network. We also find that the AWG/spl par/PSC network gives over a wide operating range a better throughput-delay performance than a network consisting of either two load sharing PSCs in parallel or two load sharing AWGs in parallel.     

Multistage shuffle networks with shortest path and deflectionrouting for high-performance ATM switching: the closed-loop shuffleout

A new class of switching architectures for broadband packet networks, called shuffleout, is described and analyzed. Shuffleout is basically an output-queued architecture with a multistage interconnection network built out of unbuffered b×2b switching elements. Its structure is such that the number of cells that can be concurrently switched from the inlets to each output queue equals the number of stages in the interconnection network. The switching element operates the cell self-routing adopting a shortest path algorithm which, in case of conflict for interstage links, is coupled with deflection routing. The basic version of this architecture is called open-loop shuffleout. This paper describes the closed-loop shuffleout architecture with 2×4 switching elements in which cells that have crossed the whole interconnection network re-enter the network as long as they are not successfully routed to the addressed switch outlet. This result is accomplished by adding to the basic open-loop structure recirculation paths so that each packet can cross several times the interconnection network. Two different solutions are proposed to implement such functionality, the buffered closed-loop shuffleout and the expanded closed-loop shuffleout architecture. Both these solutions aim at reducing the number of stages in the network, compared to the open-loop structure, so as to reduce the complexity of the switch internal wiring and to simplify the output queue interface     

Optimal physical carrier sense in wireless networks

We investigate the problem of maximizing Medium Access Control (MAC) throughput in Carrier Sense Multiple Access (CSMA) wireless networks. By explicitly incorporating the carrier sense threshold and the transmit power into our analysis, we derive an analytical relation between MAC throughput and system parameters. In homogeneous networks, we derive the optimal carrier sense range at a given node density as a function of the ratio between the transmit power and the carrier sense threshold. The obtained optimal carrier sense range is smaller than that for covering the entire interference range, which is in sharp contrast to what has been considered to be optimal in previous studies. Only when the node density goes to infinity, the optimal carrier sense range converges to that for exactly covering the interference range, thereby eliminating all the hidden nodes. For nonhomogeneous networks, any distributed algorithm for tuning the carrier sense threshold, in which each node tries to maximize its own throughput without coordination, may significantly degrade MAC throughput. In order to properly design a distributed algorithm, each node not only considers its own throughput, but also needs to take account of its adverse impact on others. Our analysis is verified by simulation studies under various network scenarios.     

物理层网络编码在数据会聚无线自组织网络中的容量增益研究

为研究物理层网络编码.(Physical-layer Network Coding,PNC).方案在数据汇聚无线自组织网络中相对于传统方案能够获得的容量增益,该文通过稀疏割容量分析推导得到PNC方案平均每节点信息传输速率的上界和下界。与传统路由方案及传统网络编码方案相比,在数据汇聚无线自组织网络中应用PNC方案能够提升网络容量的数量级。     

Bottlenecks and stability in networks with contending nodes

This paper considers a class of queueing network models where nodes have to contend with each other to serve their customers. In each time slot, a node with a non-empty queue either serves a customer or is blocked by a node in its vicinity. The focus of our study is on analyzing the throughput and identifying bottleneck nodes in such networks. Our modeling and analysis approach consists of two steps. First, considering the slotted model on a longer timescale, the behavior is described by a continuous time Markov chain with state dependent service rates. In the second step, the state dependent service rates are replaced by their long run averages resulting in an approximate product form network. This enables us to determine the bottleneck nodes and the stability condition of the system. Numerical results show that our approximation approach provides very accurate results with respect to the maximum throughput a network can support. It also reveals a surprising effect regarding the location of bottlenecks in the network when the offered load is further increased.     

Performance analysis of the IEEE 802.11e wireless networks with TCP ACK prioritization

With the advent of multimedia over wireless local area networks, the IEEE 802.11e standard was proposed to incorporate Quality of Service (QoS). It has been found that the throughput of Transmission Control Protocol (TCP) is less than that of User Datagram Protocol (UDP) in the IEEE 802.11e. This is because the TCP acknowledgment packets are queued up at the access points. In this paper, two types of TCP acknowledgment prioritizing schemes are proposed. The proposed schemes improve the overall throughput of TCP while maintaining the QoS requirements. We also analyze the problem of starvation of lower priority traffic and its effects on the performance of lower priority TCP traffic. The proposed dynamic scheme of TCP acknowledgment prioritization aims at improving the throughput of the lower priority TCP traffic under heavy network load while maintaining the QoS requirements of the higher priority traffic. The schemes have been verified through extensive simulation.     

Using the multistage cube network topology in parallelsupercomputers

A critical component of any large-scale parallel processing system is the interconnection network that provides a means for communication along the system's processors and memories. Attributes of the multistage cube topology that have made it an effective basis for interconnection networks and the subject of much ongoing research are reviewed. These attributes include O(N log₂N) cost for an N-input/output network, decentralized control, a variety of implementation options, good data-permuting capability to support single-instruction-stream/multiple-data-stream (SIMD) parallelism, good throughput to support multiple-instruction-stream/multiple-data-stream (MIMD) parallelism, and ability to be partitioned into independent subnetworks to support reconfigurable systems. Examples of existing systems that use multistage cube networks are considered. The multistage cube topology can be converted into a single-stage network by associating with each switch in the network a processor (and a memory). Properties of systems that use the multistage cube network in this way are examined