Performances of the Data Vortex switch architecture under nonuniform and bursty traffic

The Data Vortex switch architecture has been proposed as a scalable low-latency interconnection fabric for optical packet switches. This self-routed hierarchical architecture employs synchronous timing and distributed traffic-control signaling to eliminate optical buffering and to reduce the required routing logic, greatly facilitating a photonic implementation. In previous work, we have shown the efficient scalability of the architecture under uniform and random traffic conditions while maintaining high throughput and low-latency performance. This paper reports on the performance of the Data Vortex architecture under nonuniform and bursty traffic conditions. The results show that the switch architecture performs well under modest nonuniform traffic, but an excessive degree of nonuniformity will severely limit the scalability. As long as a modest degree of asymmetry between the number of input and output ports is provided, the Data Vortex switch is shown to handle very bursty traffic with little performance degradation.     

Performance evaluation of multiple input-queued ATM switches withPIM scheduling under bursty traffic

In this letter, we analyze the performance of multiple input-queued asynchronous transfer mode (ATM) switches that use parallel iterative matching (PIM) for scheduling the transmission of head-of-line cells in the input queues. A queueing model of the switch is developed under independently, identically distributed, two-state Markov modulated Bernoulli processes bursty traffic. The underlying Markov chain of the queueing model is a quasi-birth-death (QBD) chain. The QBD chain is solved using an iterative computing method. Interesting performance metrics of the ATM switch such as the throughput, the mean cell delay, and the cell loss probability can be derived from the model. Numerical results from both the analytical model and simulation are presented, and the accuracy of the analysis is briefly discussed     

Performance of two-stage shared fiber delay line optical packet switch under bursty traffic

Packet contention is a major issue in an optical packet switching network. It is not a trivial task to resolve contention due to lack of optical RAM technology. This article proposes a two-stage shared fiber delay line (FDL) optical packet switch for contention resolution. In this article, shared FDLs are used to buffer optical packets, in which a pool of buffer memory is shared among all switch output ports. Most of the existing optical buffering schemes are output-based which require a huge number of FDLs as well as a larger switch size that incur extra implementation cost. However, a shared buffering approach is considered in this article in order to reduce implementation cost. In this article, FDLs are implemented in two stages using an extremely simple auxiliary switch. The proposed switch architecture leads to more efficient use of buffer space. The superiority of the proposed switch architecture has been established by means of extensive simulations. The performance of the proposed switch is investigated under bursty traffic. Simulation result shows that the proposed switch can achieve satisfactory performance at the price of a reasonable amount of FDLs. Moreover, the significance of the proposed switch is confirmed by simulation.     

Managing bandwidth in ATM networks with bursty traffic

Three approaches to the bandwidth management problem that have been proposed and studied by various groups are reviewed to illustrate three distinctly different approaches and identify their strengths and weaknesses. Based on these approaches, a bandwidth management and congestion control scheme for asynchronous transfer mode (ATM) networks that supports both point-to-point and one-to-many multicast virtual circuits is proposed. It is shown that the method can handle fully heterogeneous traffic and can be effectively implemented. The algorithm for making virtual circuit acceptance decisions is straightforward and fast, and the hardware mechanisms needed to implement buffer allocation and traffic monitoring at the user-network interface have acceptable complexities. It is also shown, through numerical examples, that the approach can achieve reasonable link efficiencies even in the presence of very bursty traffic. No advance reservation required, simplifying the interface between the network and the user and avoiding an initial network round trip delay before data can be transmitted     

ABR performance in presence of bursty TCP traffic

The available bit rate (ABR) service class is a solution for the integration of data traffic in asynchronous transfer mode networks. Many algorithms have been proposed to implement ABR services. The authors present simulation results showing poor performance by a common ABR algorithm when supporting TCP bursty traffic. As a solution to this problem, the authors propose time averaging of the parameters calculated by the ABR algorithms     

A new analytical model for the CQ switch throughput calculation under the bursty traffic

In this paper we propose a new analytical iterative method for the throughput calculation of the Crosspoint Queued (CQ) switch with a random scheduling algorithm under the bursty traffic model. This method is verified by comparing it with the simulation results, which shows a very good match. To the authors’ knowledge, this is the first analytical method for the throughput calculation of such a switch for the bursty traffic model.     

Heterogeneous bursty traffic dispersion over multiple server clusters

This letter presents performance analysis of multiple subnets, each representing a cluster of computing server nodes, that are introduced with non-uniformly distributed bursty packet arrivals. In particular, we study the case of a multi-state Markov-modulated arrival process, heterogeneously dispersed among designated queues. Cluster processing is modeled by employing a batch service discipline. The probability generating functions of the interarrival times distributions are utilized to derive closed-form expressions for each of the queue size distributions.     

Adaptive MAC protocols for broadcast networks with bursty traffic

An adaptive medium access control protocol for broadcast networks, which is capable of operating efficiently under bursty traffic conditions, is introduced. According to the proposed protocol, the station which grants permission to transmit at each time slot is selected by taking into account the network feedback information. In this way, the number of idle slots is minimized and the network performance is significantly improved. Furthermore, the portion of the bandwidth assigned to each station is dynamically adapted to the station's needs.     

Delay-bounded packet scheduling of bursty traffic over wireless channels

In this paper, we study minimal power transmission of bursty sources over wireless channels with constraints on mean queuing delay. The power minimizing schedulers adapt power and rate of transmission based on the queue and channel state. We show that packet scheduling based on queue state can be used to trade queuing delay with transmission power, even on additive white Gaussian noise (AWGN) channels. Our extensive simulations show that small increases in average delay can lead to substantial savings in transmission power, thereby providing another avenue for mobile devices to save on battery power. We propose a low-complexity scheduler that has near-optimal performance. We also construct a variable-rate quadrature amplitude modulation (QAM)-based transmission scheme to show the benefits of the proposed formulation in a practical communication system. Power optimal schedulers with absolute packet delay constraints are also studied and their performance is evaluated via simulations.     

Learning-automata-based TDMA protocols for broadcast communicationsystems with bursty traffic

A learning automata-based time-division multiple-access protocol for broadcast networks, which is capable of operating efficiently under bursty traffic conditions, is introduced. According to the proposed protocol, the station which grants permission to transmit at each time slot is selected by means of learning automata. The learning automata update the choice probability of each station according to the network feedback information in such a way that it asymptotically tends to be proportional to the probability that this station is ready. In this manner, the number of idle slots is minimized and the network performance is significantly improved. Furthermore, the portion of the bandwidth assigned to each station is dynamically adapted to the station's needs     

Analysis of interdeparture processes for bursty traffic in ATMnetworks

Admission control in asynchronous transfer mode (ATM) networks is considered. An ATM switch is modeled as a discrete-time single-server queue at which the following three different kinds of arrival processes are allowed to join together: arrivals of cells with a general interarrival time distribution; Bernoulli arrivals of cells in batches; and interrupted Poisson processes. An exact analysis is given to derive the waiting-time distributions and interdeparture-time distributions for arriving cells subject to admission control in ATM networks. The model is extended to approximately obtain the end-to-end delay distributions for the designated traffic stream. Such an analysis is important for voice or coded video because they require a playout mechanism at the destination. Since the analysis is approximate, the simulation results needed to assess its accuracy are provided. Numerical examples show how the traffic characteristics of the designated traffic stream are affected by the total traffic load at the switching node, the number of other bursty traffic sources, and the degree of burstiness of the traffic source itself     

Tuning RED for Web traffic

We study the effects of RED on the performance of Web browsing with a novel aspect of our work being the use of a user-centric measure of performance: response time for HTTP request-response pairs. We empirically evaluate RED across a range of parameter settings and offered loads. Our results show that: (1) contrary to expectations, compared to an FIFO queue, RED has a minimal effect on HTTP response times for offered loads up to 90% of link capacity; (2) response times at loads in this range are not substantially affected by RED parameters; (3) between 90% and 100% load, RED can be carefully tuned to yield performance somewhat superior to FIFO, however, response times are quite sensitive to the actual RED parameter values selected; and (4) in such heavily congested networks, RED parameters that provide the best link utilization produce poorer response times. We conclude that for links carrying only Web traffic, RED queue management appears to provide no clear advantage over tail-drop FIFO for end-user response times     

Bandwidth allocation for bursty isochronous traffic in a hybridswitching system

Consideration is given to a hybrid switch which provides integrated packet and circuit switching. The statistical characteristics of the bandwidth utilized by bursty isochronous traffic are derived. The derivation takes into consideration the waste inherent in the framing structure. Bursty traffic due to circuit allocation requests is modeled by a switched Poisson process. The analysis is performed for two circuit allocation policies, namely, repacking and first-fit with static numbering. Numerical results are presented to demonstrate the effect of traffic burstiness on the capacity utilized by the isochronous traffic     

Supporting bursty traffic with bandwidth guarantee in WDM distribution networks

This paper presents new research results of the DARPA-funded ONRAMP consortium on the next generation Internet to study efficient WDM-based network architectures and protocols for supporting broadband services in regional access networks. In particular, we present new efficient scheduling algorithms for bandwidth sharing in WDM distribution networks. The current ONRAMP distribution network architecture has a tree topology with each leaf node (e.g., a router or workstation) sharing access to the root node of the tree, which corresponds to an access node in the feeder network. Our model allows a leaf node to use one or more fixed-tuned or tunable transceivers; moreover, different leaf nodes can support different subsets of wavelengths depending on their expected traffic volumes. An important goal of ONRAMP is to support bandwidth-on-demand services with QoS guarantee over WDM. As a first step toward this goal, we have developed several fast scheduling algorithms for flexible bandwidth reservations in a WDM distribution network. The scheduling algorithms can provably guarantee any bandwidth reservations pattern that does not overbook network resources, i.e., bandwidth reservation (throughput) up to 100% network capacity can be supported.     

Computational methods for performance evaluation of a statisticalmultiplexer supporting bursty traffic

A statistical multiplexer supporting a number of bursty sources is modeled as a discrete time, single server queueing system with an infinite buffer. The probability generating function (PGF) method is used to analyze the queueing behavior. The PGF method requires the determination of a large number of boundary values and, hence, the roots of the characteristic equation. An iterative algorithm to evaluate the characteristic roots is proposed. When the arrival process is a superposition of independent processes, a decomposition approach is used to reduce the state space involved in the computational algorithm. Additionally, the computational algorithm is made efficient through the establishment of conditions under which all the roots are either real or nonnegative real numbers. A set of equations to recursively compute the moments of the queue length are established. Sample applications of the computational methods to evaluate the performance of a multiplexer supporting voice and video sources, modeled by two-state Markov and L-state MMPP processes, respectively, demonstrate the viability of the proposed methods     

一种新的突发分配业务模型以及在OBS中的应用

文章提出了一种新的突发分配业务模型,给出了这种业务模型的详细定义,并应用此模型和传统的非突发分配模型对光突发交换(OBS)中的交换机构进行了性能分析和对比.计算结果表明,当交换机构扇出比F=1、突发强度B1=2时,突发分配模式下的丢包率比非突发分配模式下的丢包率增加大约一个数量级.也许该突发业务模型并不能真实地反映现实世界的业务流,但其能提供一种逼近现实世界业务流的分析方法.     

Buffer sizing for synchronous self-routeing broadband packet switches with bursty traffic

Previous studies on the performance of synchronous self-routeing packet switches have assumed that the input traffic is random, i.e. there is no correlation between adjacent packet arrivals. This assumption is generally not valid in the data communication environment (e.g. host-to-host communication) where a file transfer usually generates a string of correlated packets. The consequence is that the random traffic assumption greatly underestimates the buffer requirement of the switch. In this paper, we model each input traffic stream as a binary source as a first step to understand the performance of a packet switch in a bursty traffic environment. We found that, given a fixed traffic load (or switch utilization), the required buffer size increases linearly as the burstiness index (the average burst length) of the traffic increases. In addition, the required buffer size is more sensitive to the burstiness of the traffic, when the average traffic load is higher and when the packet loss requirement is more stringent. Initial applications of broadband packet switches are likely to be the interconnections of LANs and hosts. The results of the study indicate that the high burstiness in certain broadband traffic significantly reduces the allowable switch utilization, given a fixed amount of buffers. To increase the switch utilization, an appropriate congestion control mechanism needs to be implemented.     

Performance analysis of reservation media-access protocol with access and serving queues under bursty traffic in GPRS/EGPRS

Performance modeling of the contention-based reservation protocol in the general packet radio service (GPRS)/enhanced GPRS (EGPRS) under bursty traffic is of practical use in system design. Instead of using discrete event simulation, we construct an analytical model based on stochastic Petri net formalism to investigate GPRS uplink performance efficiently. The model is built to capture the major features of a realistic system, taking into account factors such as the traffic profile, the system's capture capability, the contention persistence, the multislot capability, and the use of the access queue and the round-robin virtual serving queue. As shown in numerical results, these factors have different impacts on the performance measures, such as link-control-layer frame-blocking probability, frame delay, system throughput, and packet-channel utilization. To avoid modeling and computational complexity in a GPRS/EGPRS system model that supports several mobile terminals, we make first-order approximations and use an iterative fixed-point scheme. Comparisons with discrete event simulations are conducted and show promising agreement with the proposed analytical-numerical model. We then use the model to investigate GPRS packet-data performance under different parameter settings and present numerical results, providing insights into network design and optimization.     

PERFORMANCE ANALYSIS OF LEAKY BUCKET ALGORITHM WITH BURSTY TRAFFIC INPUT IN ATM NETWORKS

A bursty traffic model is introduced in this paper to describe the statistical characteristics of packet video. The performance of leady bucket algorithm with bursty traffic input is analyzed. The influences of various parameters on QOS (Quality of Service) are investigated. The analysis shows that although the loss probability decreases through expanding the buffer capacity, the delay and delay jitter increase, whose effect on QOS will not be negligible.     

Management traffic in emerging remote configuration mechanisms for residential gateways and home devices

XML technology is penetrating the network management in the IETF, UPnP, and DSL forum suite of standards and protocols. The advantages of XML are offered at the cost of long byte streams due to XML's inherently verbose nature. The increase in packet size for remote configuration and management can pose problems, if executed in a point-to-multipoint arrangement comprising one automatic configuration server and thousands of home gateways and multimedia devices. We investigate and exploit the repetitive nature of text patterns in typical XML documents as produced by the configuration and management tasks and as coded in SOAP RPCs. The solution mainly comes from application of the Lempel-Ziv compression algorithm, with minimal additions in the proposed DSL Forum standard. Numerical and experimental results support the applicability and advantages of the proposed approach and provide insight on how these are attributable to different layers of the employed protocol stack.