Design of a non-blocking shared-memory copy network for ATM

The design of a copy network is presented for use in an ATM (asynchronous transfer mode) switch supporting BISDN (broadband integrated services digital network) traffic. Inherent traffic characteristics of BISDN services require ATM switches to handle bursty traffic with multicast connections. In typical ATM switch designs a copy network is used to replicate multicast cells before being forwarded to a point-to-point routeing network. In such designs, a single multicast cell enters the switch and is replicated once for each multicast connection. Each copy is forwarded to the routeing network with a unique destination address and is routed to the appropriate output port. Non-blocking copy networks permit multiple cells to be multicasted at once, up to the number of outputs of the copy network. Another critical feature of ATM switch design is the location of buffers for the temporary storage of transmitted cells. Buffering is required when multiple cells require a common switch resource for transmission. Typically, one cell is granted the resource and is transmitted while the remaining cells are buffered. Current switch designs associate discrete buffers with individual switch resources. Discrete buffering is not efficient for bursty traffic as traffic bursts can overflow individual switch buffers and result in dropped cells, while other buffers are under-used. A new non-blocking copy network is presented in this paper with a shared-memory input buffer. Blocked cells from any switch input are stored in a single shared input buffer. The copy network consists of three banyan networks and shared-memory queues. The design is scalable for large numbers of inputs due to low hardware complexity, O (N log₂ N), and distributed operation and control. It is shown in a simulation study that a switch incorporating the shared-memory copy network has increased throughput and lower buffer requirements to maintain low packet loss probability when compared to a switch with a discrete buffer copy network.     

Evaluation of congestion control protocols for ABR traffic over ATM networks

Congestion control is very important for effective and stable operation of ATM (Asynchronous Transfer Mode) networks. Owing to the bursty and unpredictable characteristic of data network traffic, its congestion control is particularly a challenge for network researchers and designers. The ATM Forum has recently adopted rate‐based congestion control for ABR (Available Bit‐Rate) traffic which is the service class defined for data network applications. However, there is a number of congestion control schemes prevalent. ATM Forum has decided not to specify switch behaviour for ABR traffic; this has further introduced additional ambiguity. Consequently, an evaluation and comparison of the existing protocols would provide valuable guidance for network designers and engineers; it would also give insight for researchers to explore the essence of different congestion control schemes. In the first part of this paper, we investigate the effectiveness of ABR congestion control in the presence of bursty source traffic and the relationship between the burst time scale and the ABR control time scale. Two ABR congestion control schemes, the ABR Explicit Forward Congestion Indication (EFCI) and ABR Congestion Indication (CI) schemes, are compared with Unspecified Bit Rate (UBR) transport which makes no effort to control congestion. Traffic sources of various burst lengths of 100, 1000, 10000, and an equal mix of 100 and 10000 ATM cells are used in simulations. It is found that ABR congestion control schemes effectively control low frequency, medium to long‐term traffic load transients. This is further supported by the result of integrating TCP over ATM congestion control schemes included in the paper. ABR control schemes do not control high frequency, short‐term load transients well, but ABR control is not necessary in such cases since short‐term transients do not require a large amount of buffering. In the second part of this paper, we evaluate and compare six rate‐based congestion control protocols including Scheme I: EFCI, Scheme II: EFCI with separate RM queues, Scheme III: CI, Scheme IV: CI with separate RM queues, Scheme V: the CAPC2 ER (Explicit Rate), and Scheme VI: the EFCI with utilization‐based congestion indication. Each scheme is simulated and compared in the LAN, WAN, and GFC (General Fairness Configuration) environments specified by the ATM Forum. Effects of varying VC (Virtual Circuits) number and changing endsystem–switch distance has been investigated. Their fairness is also compared using the GFC configuration. We have found that ER control scheme performs significantly better than the other five binary control schemes by its faster response to congestion, smoother regulation of bit‐rates, lower queueing delay, shorter buffer queue length, and fairness. Among the other five schemes, the CI scheme performs better than the EFCI scheme. Providing separate RM queues has significantly improved the EFCI scheme in the WAN environment, but has little effect on the CI scheme. Link utilization‐based congestion detection has suffered from either low utilization or an excess cell loss which is unacceptable in most data applications. Copyright © 2000 John Wiley & Sons, Ltd.     

Distributed source control: a network access control for integratedbroadband packet networks

The authors propose a congestion control strategy called distributed source control (DSC), designed to address the integration of diverse traffic types in broadband packet networks. DSC is a rate-based network access control implemented at the asynchronous transfer mode (ATM) layer as a feedforward control, in conjunction with an adaptive end-to-end control between network edges. The authors examine the performance of high-speed data traffic such as large file transfers, still images, and document retrievals. The authors study DSC's integrability with real-time traffic such as voice and video. With the help of a simulation model of a broadband packet network (150 Mb/s), the authors quantify the improvement in network performance due to DSC. An analytic model for an access node under DSC is developed, and guidelines for sizing of buffers in the switch and for choosing the end-to-end window size are provided     

A simplified performance evaluation for bursty multiclass trafficin ATM systems

In broadband ISDN with ATM, various kinds of traffic, such as voice, data and video, share the common transmission resource. For such multiclass traffic systems, it is necessary to evaluate the grade of service (GOS) for individual traffic. In particular, for the ATM networks integrating voice and video which are of bursty property and delay sensitive, the individual performance evaluation for the mixed bursty traffic an important problem. The authors propose a simplified evaluation of the individual mean waiting time for the bursty multiclass systems appearing in the ATM network and show numerical examples for voice/video system     

Resource allocation for broadband networks

The major benefit of a broadband integrated ATM (asynchronous transfer mode) network is flexible and efficient allocation of communications bandwidth for communications services. However, methods are needed for evaluating congestion for integrated traffic. The author suggests evaluating congestion at different levels, namely the packet level, the burst level, and the call level. Congestion is measured by the probabilities of packet blocking, burst blocking, and call blocking. He outlines the methodologies for comparing these blocking probabilities. The author uses the congestion measures for a multilayer bandwidth-allocation algorithm, emulating some function of virtual circuit setup, fast circuit switching, and fast packet switching at these levels. The analysis also sheds insight on traffic engineering issues such as appropriate link load, traffic integration, trunk group and switch sizing, and bandwidth reservation criteria for two bursty services     

Resource allocation and connection admission control in satellitenetworks

This work focuses on resource allocation and connection admission control (CAC) issues in broadband satellite networks. Broadband services can now be provided by satellite systems operating in the Ka band, due to the large bandwidth available at such frequencies. In this context, we propose a resource allocation algorithm which integrates three classes of services at the MAC layer: constant bit rate (CBR), bursty data, and best effort services. The double movable boundary strategy (DMBS) is proposed to establish a resource-sharing policy among these service classes over the satellite uplink channel. The DMBS is a dynamically controlled boundary policy which adapts the allocation decision to the variable network loading conditions. Connection-oriented and connectionless services can be supported by the system. The CAC and slot allocation decisions are taken at the beginning of each control period after monitoring the filling level of traffic request queues. A threshold level for the bursty data request queue is defined to regulate the CAC process. The impact of the queue threshold value on the performance of the DMBS allocation policy is particularly evaluated in this study. A dynamic variation of this value is also proposed to enhance the system response to interactive applications. We present a brief analytical formulation for the DMBS model, together with simulation study details and performance evaluation results. The obtained results indicate a good efficiency, in terms of overall channel throughput and CBR blocking probability, for both fixed and dynamic data queue threshold approaches. The dynamic approach, however, outperforms the fixed one in terms of overall encountered bursty data delay     

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 evaluation of high speed TCP over optical burst switching networks

Study of TCP performance over OBS networks has been an important problem of research lately and it was found that due to the congestion control mechanism of TCP and the inherent bursty losses in the Optical Burst Switching (OBS) network, the throughput of TCP connections degrade. On the other hand, High Speed TCP (HSTCP) was proposed as an alternative to the use of TCP in high bandwidth-delay product networks. HSTCP aggressively increases the congestion window used in TCP, when the available bandwidth is high and decreases the window cautiously in response to a congestion event. In this work, we make a thorough simulation study of HSTCP over OBS networks. While the earlier works in the literature used a linear chain of nodes as the network topology for the simulation, we use the popular 14-node NSFNET topology that represents an arbitrary mesh network in our study. We also study the performance of HSTCP over OBS for different bandwidths of access networks. We use two different cases for simulations where in the first HSTCP connections are routed on disjoint paths while in the second they contend for resources in the network links. These cases of simulations along with the mesh topology help us clearly distinguish between the congestion and contention losses in the OBS network and their effect on HSTCP throughput. For completeness of study, we also simulate TCP traffic over OBS networks in all these cases and compare its throughput with that of HSTCP. We observe that irrespective of the access network bandwidth and the burst loss rate in the network, HSTCP outperforms TCP in terms of the throughput and robustness against multiple burst losses up to the expected theoretical burst loss probability of 10⁻³.     

光突发交换网络中突发组装办案研究

光突发交换（Optcal Burst Switching.OBS）被认为是在下一代光网络中起到至关重要作用的一种交换技术。边缘节点处突发包的组装作为业务接入OBS网络的第一步,是OBS研究的关键课题之一。 该文重点阐述了几种常用的突发组装方案：基于突发长度组装方案、基于组装时间组装方案、混合组装方案及自适应组装方案,并分析了它们的性能和优缺点。最后指出综合考虑网络业务流量、边缘节点的处理速度和信道带宽等因素的更加有效的突发组装算法是重点研究的内容。     

Broadband satellite networks-the global IT bridge

Next-generation broadband satellite networks are being developed to carry bursty Internet and multimedia traffic in addition to the traditional circuit-switched traffic (mainly voice) on a global basis. These satellites provide direct network access for personal applications as well as interconnectivity to the terrestrial remote network segments. The main requirement in success of these networks is that they should be able to transmit high data rate traffic with prescribed quality of service (QoS). Thus, the broadband satellite network has no choice other than the rise of ATM technology and to be optimized for Internet-based traffic. ATM is the promising technology for supporting high-speed data transfer potentially suitable for all varieties of private and public telecommunications networks. IP, on the other hand is the fast-growing Internet layer protocol that is applicable over any data link layer Internet-based applications are the emerging source of traffic in the future wireless networks and broadband satellite networks should consider Internet as the primary service. In this paper, we discuss the traditional ATM and wireless ATM networks and explain the characteristics of the wireless IP networks. The paper then uses those concepts in defining the criteria for the broadband satellite networks such as the QoS and traffic management. Application of the broadband satellite networks is also proposed     

Analysis of GSM/GPRS Cell with Multiple Data Service Classes

This paper is concerned with the performance modelling and evaluationof a General Packet Radio Service (GPRS) for a wireless Global System forMobile telecommunication (GSM) cellinvolving both voice and multiple class data services under a completepartitioning scheme (CPS). The GPRS is modelled as a system ofaccess and transfer queues in tandem having an external multiple classcompound Poisson bursty traffic with geometrically distributedbatches under first-come-first-served (FCFS) and generalised processor-sharing(PS) scheduling disciplines, respectively. The proposed queueing model isvalid for both uplink and downlink connections and provides multiple classdata services with different arrival rates, interarrival-time squaredcoefficients of variation (SCVs), file (burst) sizes and different PSdiscrimination service levels.Based on the principle of maximum entropy (ME), a product formapproximation is characterised for the two queues in tandem, subject to queueing theoretic mean value constraints. The MEsolution implies a decomposition of the system into individual singleserver building block queues, each of which can be analysed inisolation. Consequently, ME closed form expressions and simulation resultsfor the aggregate state probability together with the marginal mean queuelength and blocking probability are divised to assess the credibility of theME solution and also to study the effect of external GPRS bursty traffic uponthe performance of the wireless cell.     

A comparative study of fuzzy versus “fixed” thresholdsfor robust queue management in cell-switching networks

High-performance cell-based communications networks have been conceived to carry asynchronous traffic sources and support a continuum of transport rates ranging from low bit-rate to high bit-rate traffic. When a number of bursty traffic sources add cells, the network is inevitably subject to congestion. Traditional approaches to congestion management include admission control algorithms, smoothing functions, and the use of finite-sized buffers with queue management techniques. Most queue management schemes, reported in the literature, utilize “fixed” thresholds to determine when to permit or refuse entry of cells into the buffer. The aim is to achieve a desired tradeoff between the number of cells carried through the network, propagation delays of the cells, and the number of discarded cells. While binary thresholds are excessively restrictive, the rationale underlying the use of a large number of priorities appears to be ad hoc, unnatural, and unclear. The paper introduces the notion of cell-blocking, wherein a fuzzy thresholding function, based on Zadeh's (1965) fuzzy set theory, is utilized to deliberately refuse entry to a fraction of incoming cells from other switches. The blocked cells must be rerouted by the sending switch to other switches and, in the process, they may incur delays. The fraction of blocked cells is a continuous function of the current buffer occupancy level unlike the abrupt. The fuzzy cell-blocking scheme is simulated on a computer. Fuzzy queue management adapts superbly to sharp changes in cell arrival rates and maximum burstiness of bursty traffic sources     

New assembly techniques and fast reservation protocols for optical burst switched networks based on traffic prediction

We propose new burst assembly schemes and fast reservation (FR) protocols for Optical Burst Switched (OBS) networks that are based on traffic prediction. The burst assembly schemes aim at minimizing (for a given burst size) the average delay of the packets incurred during the burst assembly process, while the fast reservation protocols aim at further reducing the end-to-end delay of the data bursts. The burst assembly techniques use a linear prediction filter to estimate the number of packet arrivals at the ingress node in the following interval, and launch a new burst into the network when a certain criterion, different for each proposed scheme, is met. The fast reservation protocols use prediction filters to estimate the expected length of the burst and the time needed for the burst assembly process to complete. A Burst Header Packet (BHP) packet carrying these estimates is sent before the burst is completed, in order to reserve bandwidth at intermediate nodes for the time interval the burst is expected to pass from these nodes. Reducing the packet aggregation delay and the time required to perform the reservations, reduces the total time needed for a packet to be transported over an OBS network and is especially important for real-time applications. We evaluate the performance of the proposed burst assembly schemes and show that a number of them outperform the previously proposed timer-based, length-based and average delay-based burst assembly schemes. We also look at the performance of the fast reservation (FR) protocols in terms of the probability of successfully establishing the reservations required to transport the burst.     

Multimedia traffic management principles for guaranteed ATM networkperformance

A framework is presented for the specification of performance requirements, and several high-level design principles for performance-driven ATM traffic controls based on resource reservation are developed. Simulation results illustrate that to guarantee performance levels, bursty traffic should be statistically multiplexed only if connection peak rates are low relative to the network link speed or burst durations are short. A pragmatic traffic management approach that favors simplicity and robustness in the traffic control design rather than optimizing bandwidth efficiency is stressed     

A two-moment performance analysis of optical burst switched networks with shared fibre delay lines in a feedback configuration

Fibre delay lines (FDLs) can substantially reduce the burst loss in Optical Burst Switching (OBS) networks and share-per-node FDL configurations can provide a more cost-efficient solution compared to architectures where delay lines are shared per port. Nevertheless, mathematical performance analysis of this configuration is more difficult due to traffic correlations arising from the shared resource. In this paper, an approximate two-moment traffic model is developed for quantifying end-to-end burst blocking probability in networks of OBS switches with share-per-node FDLs. The two-moment approach can improve model accuracy over more usual Poisson network analysis methods and additionally allows the characteristics of offered load to be taken into account. The accuracy of the proposed method is shown to be favourable, when compared to discrete-event simulations of an OBS network.     

An efficient method for computing cell loss probability for heterogeneous bursty traffic in ATM networks

In this paper, we present an approximation for the probability of cell loss of heterogeneous bursty traffic in broadband integrated packet networks based on the asynchronous transfer mode. The sources considered here alternate between active and silent periods and are characterized by their peak and average transmission rates. The cell loss probability is obtained by considering only the number of active sources at a given time and computing the amount of traffic that exceeds the link capacity. Since the amount of buffered excess traffic is not considered in this computation, this approximation is actually an upper bound of the cell loss probability. The numerical efficiency of this bound enables it to be used as a measure based on which robust and simple resource allocation strategies can be developed for bursty sources. Comparison of this bound with the simulation results has shown that the bound is close to the actual loss probabilities especially for large burst lengths and high utilizations.     

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.     

PERFORMANCE ANALYSIS FOR ATM SWITCHING OF MIXED CONTINUOUS-BIT-RATE AND BURSTY TRAFFIC WITH SMOOTHING FUNCTION

Traditional packet switching networks have typically employed window-based congestion control schemes in order to regulate traffic flow. In ATM networks, the high speed of the communication links and the varied nature of the carried traffic make such schemes inappropriate. Therefore, simpler and more efficient schemes have to be proposed to improve the congestion control for ATM switching. This paper presents an exact performance analysis of ATM switching whose inputs consist of Continuous-Bit-Rate(CBR) and bursty traffic. The CBR traffic and bursty traffic are described by Bernoulli process and the Interrupted Bernoulli Process(IBP), respectively. Bursty traffic smoothing mechanism is analyzed. With the use of a recursive algorithm, the cell loss probability and the average delay for ATM switching of mixed CBR and bursty traffic are exactly calculated. Traffic smoothing could be implemented at a slower peak rate keeping the average rate constant or decreasing the average bursty length. Both numerical a     

Fair internet traffic integration: network flow models and analysis

We use flow-level models to study the integration of two types of Internet traffic, elastic file transfers and streaming traffic. Previous studies have concentrated on just one type of traffic, such as the flow level models of Internet congestion control, where network capacity is dynamically shared between elastic file transfers, with a randomly varying number of such flows. We consider the addition of streaming traffic in two cases, under a fairness assumption that includestcp-friendliness as a special case, and under certain admission control schemes. We establish sufficient conditions for stability, using a fluid model of the system. We also assess the impact of each traffic type on the other: file transfers are seen by streaming traffic as reducing the available capacity, whereas for file transfers the presence of streaming traffic amounts to replacing sharp capacity constraints by relaxed constraints. Simulation results suggest that the integration of streaming traffic and file transfers has a stabilizing effect on the variability of the number of flows present in the system.     

A new buffer management scheme for hierarchical shared memoryswitches

We study a multistage hierarchical asynchronous transfer mode (ATM) switch in which each switching element has its own local cell buffer memory that is shared among all its output ports. We propose a novel buffer management technique called delayed pushout that combines a pushout mechanism (for sharing memory efficiently among queues within the same switching element) and a backpressure mechanism (for sharing memory across switch stages). The backpressure component has a threshold to restrict the amount of sharing between stages. A synergy emerges when pushout, backpressure, and this threshold are all employed together. Using a computer simulation of the switch under symmetric but bursty traffic, we study delayed pushout as well as several simpler pushout and backpressure schemes under a wide range of loads. At every load level, we find that the delayed pushout scheme has a lower cell loss rate than its competitors. Finally, we show how delayed pushout can be extended to share buffer space between traffic classes with different space priorities