Performance evaluation of prioritized scheduling with buffer management for differentiated services architectures

This paper analyses single server correlated batch arrival and correlated batch service queue with innumerable waiting space in discrete-time. The arrival and service processes are governed by discrete batch Markovian arrival process and discrete Markovian service process with general bulk service rule, respectively. Observing two consecutive random epochs, we construct general structure transition probability matrix and it is then reblocked to the desired M/G/1 structure. The UL-type RG-factorization approach combined with spectral expansion method is adopted to unravel the queue length distribution at random epoch. The relationships are undertaken to obtain the queue length distributions at pre-arrival, outside observer's, intermediate and post-departure epochs. Our most challenging effort is to derive the probability mass functions of waiting time and service batch size for arbitrary customer in an arriving batch. We address the warehouse management process for finished products from dealer to retailer as a potential practical use of our batch queue with correlation in both arrival and service, and provide the effect of our queueing behaviour in this application based on our analytical and computational investigation. Based on the implementation of parametrized numeric experiments with numerous categories, we impart our computational results to validate the analytical findings reported in this investigation.

Recent advances in high-speed wireless LANs with physical layer (PHY) rates reaching 600Mbps make them ideal for multimedia applications. It has been shown that efficiency at medium access control (MAC) layer decreases with increasing the PHY rate. To achieve high efficiency, few researches have tried to use aggregation in which few packets are concatenated into a larger frame. The resultant frame is sent at once in order to reduce the media access protocol overhead. Unfortunately larger frames increase delays. Since demand for real-time data communication and delay-sensitive applications have become more critical, especially for high-speed networks, a thorough delay analysis is required to examine the effect of aggregation on delay in high-speed networks. In this paper, we propose an analytical model for examining packet delay in IEEE 802.11n networks in the saturation mode. To the best of our knowledge, this is the first work which focuses on packet delay analysis in the presence of aggregation. Analysis and simulation results show that packet delay increases in larger aggregation sizes, and IEEE 802.11n performs poorly in high rates in terms of packet delay. Moreover, based on two new metrics that we propose in the paper, it is shown that aggregation loses its effectiveness in higher rates.

This paper presents the performance evaluation of per-hop forwarding behavior in the differentiated services Internet. Both the expedited forwarding and the assured forwarding behaviors are observed and investigated in terms of packet delay and packet drop rate. This investigation was carried out using Queen's University IP Simulator v2.0 (QUIPS-II). A significant amount of results have been obtained under various parameter settings. The results show that the Diffserv per-hop forwarding mechanisms can provide good service differentiation on the Internet.

The explosive growth of the high-speed multimedia networks and the widespread use of web-based related applications place new demands on the network servers. The network end systems such as PC-Router, network server and host connected to high speed links must satisfy QoS requirements for multimedia traffics such as delay and loss ratio. Most works on operating systems support for high-speed network interface have focused in increasing the throughput and decreasing the interrupt handing overhead. However, the problem is that not all traffic streams are equal in terms of QoS requirements. Multimedia applications often depend more on low-latency than on high throughput. This study first compares the performance measures of hard timer and soft timer polling schemes used for high-speed network interface with high traffic load. Then it modifies the polling-based interrupt handling for high-speed network to not just eliminate the interrupt overhead but also to guarantee the QoS requirements for multimedia traffic. Meanly we propose an input network interface mechanism, which combines the advantages of using polling interrupt handling under high traffic load and using multi-priority queues scheduling algorithm to provide the QoS requirements. In addition to the throughput performance metric in which most of the literatures focus on only, other performance metrics such as CPU availability, loss ratio, packet delay are defined and studied. The performance evaluations, which are performed using a discrete event simulation, indicate that under conditions of high traffic load, the proposed system offers increased throughput and reduced latency for real time traffics.

This paper presents analyses to study the persession performance of an ATM multiplexer which finds application in call admission control (CAC). We consider a scenario where a switching node is non-blocking and output buffered, links are of different speeds and the traffic is bursty with correlated arrivals. We model the session of interest as a Markov modulated Bernoulli process (MMBP) and the interfering traffic as an MMBP with batch arrivals. We propose an efficient method to obtain the exact steady-state performance results of a session. An approximation technique is also presented. This method could also be used to analyze any Markov renewal traffic model such as discrete-batch Markovian arrival process (D-BMAP), and study the transient behavior of a session.

Finding a document or resource in an unstructured peer-to-peer network can be an exceedingly difficult problem. In this paper we propose a query routing approach that accounts for arbitrary overlay topologies, nodes with heterogeneous processing capacity, e.g., reflecting their degree of altruism, and heterogenous class-based likelihoods of query resolution at nodes which may reflect query loads and the manner in which files/resources are distributed across the network. The approach is shown to be stabilize the query load subject to a grade of service constraint, i.e., a guarantee that queries' routes meet pre-specified class-based bounds on their associated a priori probability of query resolution. An explicit characterization of the capacity region for such systems is given and numerically compared to that associated with random walk based searches. Simulation results further show the performance benefits, in terms of mean delay, of the proposed approach. Additional aspects associated with reducing complexity, estimating parameters, and adaptation to class-based query resolution probabilities and traffic loads are studied.

In Vehicular ad-hoc networks (VANETs), resource allocation has become one of the primary tasks. Thus, a system should be well-designed so that information could be transmitted reliably and efficiently. As vehicular communication has several characteristics, efficient resource allocation scheme for D2D communications is highly required in VANET. The network topology changes rapidly as the vehicles are moving fast on the road and this makes it harder to communicate among the vehicles by predicting the best available channel. So as to overcome these issues, the proposed technique is offered in which the channel availability is predicted based on the novel mechanism like contention latency prediction using interframe spaces like Short interframe space, Distributed Coordination Function interframe space, Pointed Coordination Function and scattering based greedy weight matrix method. The implementation of this greedy mechanism in turn enhances the prediction probability of best channels that are available. Thus, the best available channel offered using greedy weight matrix scheme thereby estimates the decision probability at which the collision for available channel is computed. Thus, from this the best available channel is provided to the requested user. At last, the performance analysis is made in terms of packet delivery ratio, throughput, delay, reliability, and rate of message loss and is compared with existing techniques to prove the effectiveness of proposed strategy.

Bluetooth is a short-range TDD (time division duplex) wireless network that supports both circuit- and packet-oriented applications. A piconet is composed of a device configured as master and at most seven other devices acting as slaves. At medium access control (MAC) layer, the master can select a slave to send a data packet and until then, the slave is not allowed to transmit. Round-robin (RR) and exhaustive round-robin (ERR) are two elementary MAC scheduling schemes that are both simple and efficient. This paper proposes RR-FCFS, a simple MAC scheduling scheme that has the same advantages as RR and ERR. RR-FCFS acts as RR if the master's queue is empty and starts transmitting packets in first-come-first-serve order otherwise. The simulation results show that RR-FCFS's performance in terms of packet delay and queue length is comparable with those of RR and ERR.

A reliable, scalable, and low-delay information collection network is an essential component in cyber-physical systems (CPS). Cluster-based sensor network is a good candidate because of its advantage in increasing scalability, improving energy efficiency, and providing QoS guarantees. However, in such networks, frequent interactions between the intra-cluster communication and the inter-cluster communication are inevitable, which may severely downgrade the communication efficiency and hence the network performance if not handled properly. Proper synchronization among these two types of communications is required. In this paper, we propose two approaches to schedule the communications in clustered wireless sensor networks aiming at delay-sensitive applications. In the first approach, an efficient cycle-based synchronous scheduling is proposed to achieve low average packet delay and high throughput by optimizing the cycle length and transmission order. In the second approach, a novel clustering structure is introduced to eliminate the necessity of communication synchronization so that packets are transmitted with no synchronization delay, yielding very low end-to-end packet delay. Our extensive experimental results demonstrate the superior performance of both the approaches. These two approaches are then integrated as a hybrid scheme that allows smooth switching between them. The hybrid scheme takes advantage of both the approaches and enables cluster-based sensor networks to serve as the fundamental network infrastructure for information collection in CPS.

Wireless Local Area Networks (WLAN)s are today common for access to Internet; well-known and used is IEEE 802.11 standardized technology. Some drawbacks of the standardized distributed medium access scheme, Distributed Coordination Function (DCF) has resulted in considerations made to extend the DCF from single-hop communications to multi-hop. Multihopping has the potential to reduce energy consumption and increase throughput and coverage. The drawback is lack of Quality-of-Service (QoS) support, which already with single-hop communication was bad, but extending a communication between two peers to multi-hop significantly reduces the possibility to offer reliable transmissions. The contention-based medium access performed at each hop with the DCF results in large uncertainties as to when a packet arrives to its end-destination. There are two general approaches for extending the DCF with QoS functionality: 1. Allocate resources for a flow and 2. Assign a flow higher priority than other flows. My hypothesis is that for strict QoS guarantees, a reservation-based approach is the better one. This has been my motivation to design a new medium access protocol; based on the DCF, periodic time slots for QoS-demanding real-time applications are reserved. The Distributed Allocation of time slots for Real-time traffic (DARE) is unique as it totally distributed sets up an end-to-end reservation before the transmission of data begins, repairs broken reservations, support many periods and time slot sizes and distribute piggy-backed information even to a two-hop radius from a receiver. This thesis describes the DARE protocol and also presents results from comparisons of DARE, DCF and the IEEE 802.11 E standardized priority-based QoS medium access protocol Enhanced Distributed Channel Access (EDCA). Using simulations, this thesis shows that DARE offers constant end-to-end packet delay for a flow, very low average packet delay in the whole system and constant throughput. EDCA and DCF depend strongly on the total network load; average packet delay for a real-time flow increases rapidly with the surrounding load. DARE outperforms both DCF and EDCA when many flows are present, even in networks with frequent topology changes where reservations must be repaired. Thus, the reservation-based DARE protocol is the most suitable approach for extending DCF with QoS.

The provision of multimedia services via a non-geostationary satellite system requires advanced traffic engineering procedures. It includes adaptive routing, adaptive forwarding and appropriate signalling. In this study, we evaluate the impact of different signalling procedures on the performance of a network. The signalling procedures were used in conjunction with traffic class dependent routing, smoothed link-cost functions and adaptive forwarding alleviates the oscillations in the traffic distribution. The performance of the proposed signalling procedures is evaluated in two different traffic load scenarios. The simulation results are presented in terms of the average packet delay in the network and the average normalized data throughput. The simulation results indicate that the proposed triggered signalling significantly decreases the signalling load, while the network's performance is not deteriorated in terms of packet delay and data throughput.

Network firewalls act as the first line of defense against unwanted and malicious traffic targeting Internet servers. Predicting the overall firewall performance is crucial to network security engineers and designers in assessing the effectiveness and resiliency of network firewalls against DDoS (Distributed Denial of Service) attacks as those commonly launched by today's Botnets. In this paper, we present an analytical queueing model based on the embedded Markov chain to study and analyze the performance of rule-based firewalls when subjected to normal traffic flows as well as DoS attack flows targeting different rule positions. We derive equations for key features and performance measures of engineering and design significance. These features and measures include throughput, packet loss, packet delay, and firewall's CPU utilization. In addition, we verify and validate our analytical model using simulation and real experimental measurements.

Virtual Output Queuing is widely used by highspeed packet switches to overcome head-of-line blocking. This is done by means of matching algorithms. In fixed-length VOQ switches, variable-length IP packets are segmented into fixed- length cells at the inputs. When a cell is transferred to its destination output, it will stay in the reassembly buffer and wait for the other cells of the same packet before the entire packet can depart the system. The delay a packet suffers in the system includes the waiting time in the VOQ, the widely studied cell delay, and the waiting time at the output reassembly buffer, the reassembly delay often ignored in many papers. Among all existing matching algorithms, Maximum Weight Matching (MWM) has the lowest average cell delay. In this paper, we investigate the average packet delay, one of the key performance measure for an input buffered packet switch. A new class of matching algorithms, PDA-MWM, is defined and proved to be stable under all admissible traffic. Three PDA-MWM matching algorithms are studied by simulation. We show that, in order to achieve low packet delay, there is a tradeoff between the cell delay performance and the reassembly delay performance. If both of them are carefully considered, a matching scheme can greatly reduce the packet delay as compared to MWM.

We propose a dynamic bandwidth allocation algorithm for ethernet passive optical networks (EPON) called IP-SLA (interleaved polling with SLA) to offer quality of service (QoS) in the upstream direction. IP-SLA differs clients inside the network as it takes into consideration the service level agreement (SLA) contracted by each subscriber, achieving better performance for the highest service levels agreement in terms of both mean packet delay and packet loss rate. On the other hand, IP-SLA also provides service differentiation based on the strict priority queueing method, which provides the minimum mean packet delay for the most priority services without considering the penalization caused to the lowest priority traffic. However, our algorithm keeps the mean packet delay for the most priority classes of services inside the specified bounds and uses the remaining resources to improve the performance of the lowest priority services in terms of mean packet delay and packet loss rate.

This paper investigates the performance of Beyond 3G system following the requirements defined in WINNER (phase I and II) Project and concentrates on two major topics. The first one is a novel frequency domain mutual-information based adaptive coding and modulation (MI-ACM), the second one is the impact of traffic modeling and scheduling on delay and spectral efficiency of such systems. Different MIMO schemes, including Alamouti and PARC utilizing Grid of Beams and SDMA, are examined on the system-level in terms of spectral efficiency and packet delays.