Traffic Shaping

Se realiza un análisis del entorno de Calidad de Servicio actual, para identificar como se puede aplicar en la red más grande que es Internet. El documento se encuentra dividido en dos partes: en la primera de ellas se hace una revisión rápida de conceptos básicos de QoS como Arquitecturas, Mecanismos de Encolamiento y Algoritmos de Regulación de Velocidad; necesarios para entender la segunda parte. En la segunda parte se hace un seguimiento del artículo “Teorías y Modelos para Calidad de Servicio en Internet”, donde se analizan conceptos modernos como Network Calculus y la propuesta de Ion Stoica para un Stateless Core; si es conocedor del tema, podría iniciarse directamente por aquí. El objetivo final del documento es bajar la teoría general de QoS a un nivel más comprensible.

Distributed Denial of Service (DDoS) attack mitigation systems usually generate a list of filter rules in order to block malicious traffic. In contrast to this binary decision we suggest to use traffic shaping whereas the bandwidth limit is defined by the probability of a source to be a legal user. As a proof of concept, we implemented a simple high performance Linux kernel module nf-HiShape which is able to shape thousands of source IP addresses at different bandwidth limits even under high packet rates. Our shaping algorithm is comparable to Random Early Detection (RED) applied on every single source IP range. The evaluation shows, that our kernel module can handle up to 50,000 IP ranges at nearly constant throughput whereas Linux tc already decreases throughput at about 200 ranges.

Timing has a key role in several traffic control functions encountered in modern packet-switched networks. In order to be effective, a timing unit must provide fine resolution, be simple to implement and scale well with the number of controlled traffic streams. This paper addresses the design, implementation and evaluation of a timing unit that can support accurate and efficient implementations of traffic shaping, policing and charging in packet-switched networks. The timing unit is implemented in hardware and, therefore, overcomes constraints associated with software-based timers. It accommodates a pool of independently-clocked timers and counters, organised in timing blocks, and, consequently, is able to support, in parallel, traffic streams with diverse timing requirements. The design supports shaping and policing through token buckets, leaky buckets and a scheme, variation of the token bucket, that aims at providing statistical quality of service guarantees by exploiting the effective rate concept. Charging is supported by dedicated counters that measure the utilization of the effective rate. The granularity of the timing unit is adjustable in run-time to adapt to changes in the rate parameters of the shaping and policing functions. The validation of the timing unit is done through the development of a prototype board consisting of programmable hardware and embedded software blocks. The temporal resolution of the timing unit and the advantages of the hardware/ software co-design are experimentally evaluated.

Media data are variable-bit-rate (VBR) in nature due to the coding and compression technologies applied. As VBR streams are complicated for network management, different approaches were proposed to shape the VBR stream as a transmission schedule with smoothed traffic burst. In this paper, instead of giving a fixed schedule result, a novel traffic shaping scheme is proposed to decide a schedulable region for all optimal transmission schedules that provides the minimal allocation and maximal utilization of system resources (such as network bandwidth, initial delay and client buffer). Experiments have shown that our obtained shaping results show dramatic improvements than that of conventional approaches in both the client buffer size and the network idle rate achieved. Based on the schedulable region provided, the ready time and deadline for each media packet can be precisely specified to support real-time network scheduling and error control. It allows users to determine their own optimal schedules under various quality-of-service (QoS) requirements and resource constraints.

Services architecture hinges on the specification of appropriate traffic control mechanisms. However, recent and past studies of Internet traffic have shown that it is rather hard to find a global and representative traffic model, which at the same time remains simple enough to facilitate the implementation of such mechanisms. In some previous works, we have demonstrated that suitable traffic shaping of packet flows that are subsequently multiplexed at a network node can permit the application of the M/G/1 model for performance modelling and control of the aggregate traffic stream. Using some handy control laws derived from the M/G/1 model, quantitative QoS guarantees can be provisioned end-to-end, across a series of multiplexers. In this paper we present our experiences from embedding this traffic-shaping scheme, which we call exponential token bucket filtering (etbf), in routers based on the Linux operating system. Initially, we describe how the traffic-shaping scheme can fit in a Differentiated Services architecture as a mechanism applied on the traffic transmitted from the edge routers to the core of a DiffServ domain, so that traffic multiplexed at the core routers conforms to the M/G/1 model. The Linux traffic control architecture is then presented, which enables the implementation of a DiffServ framework as well as how this shaping scheme is implemented within this architecture. Finally, results from experiments in a lab environment consisting of such Linux edge routers are presented. These results are in-line with findings from experiments and simulations in our previous works and show that the performance at the core multiplexing nodes can be controlled according to the M/G/1 model. q

This paper presents a VLSI architecture specifically designed as a video/communication controller to support emerging applications in the area of video/data communications. The controller is a parallel architecture consisting of three (3) processing modules, a shared memory with four (4) banks and two (2) input/output modules and operating at the transfer speed of 622 Mbits/sec. The processing modules and memory banks communicate through a low cost interconnection scheme able though to perform at system's required data transfer rate. The entire system constitutes a component which can accommodate a switching system as an intelligent buffer with real time processing and multiplexing capabilities. The component performs operations on fixed and/or variable length packets of data on a stream basis. The architecture embeds both the processing and the memory modules, thus producing a “system on a chip” solution.

One of the causes of congestion in computer networks is the inherent burstiness of network traffic. Traffic includes unpredictable file or web page transfers as well as variable rate media streams. Traffic shaping means to smooth the traffic as a function of time. Shaping and congestion control is very important especially in cases where Variable Bit Rate (VBR) traffic streams are passing through the network. Many different schemes to address this issue have been proposed. The target of such schemes is to maximize the amount of the traffic that can be supported while maintaining the required quality of service and to reduce the amount of congestion in the network. This paper investigates the effect of using various shapers on VBR streams on the overall diffserv domain. Simulations are conducted using NS-2 simulation platform applying VBR streams that are shaped using TB and dual-TB shapers followed by srTCM, trTcm and TSWTCM markers. The shaped traffic is applied across a diffserv d...

We evaluate and compare the performance of three P2P streaming systems that are capable of streaming live video on the Internet by testing them on a carefully controlled, traffic-shaped network testbed. We first describe the construction of the test-bed based upon Internet measurements between geographically distributed hosts. Then, we present a methodology for evaluating these P2P video streaming systems by performing video quality and network usage analysis from the log information obtained via running these systems on the test-bed. Our methodology to assess P2P live video streaming systems comprises analyzing the objective quality of the received video, waiting time to receive the first data byte, and several network usage measures such as P2P protocol overhead, load on the server due to the inefficiencies of the P2P overlay, and measurements of the number of bytes exchanged between the peers. It is essential that every peer buffers packets for some time before playing out the video in order to ensure good quality. We report the time that the user has to wait before he can see the video playing. These measurements not only gauge the performance of currently available P2P streaming systems but also highlight desired improvements in current P2P video streaming systems.

This paper concerns the modelling of Internet packet traffic. In previous work we showed that a Bartlett-Lewis point process, as a model of packet arrivals on backbone links, enjoys strong physical backing and can predict key features. It is based on the surprising empirical observation that flows can often be considered independent for the purpose of modelling packet arrival times. We extend this work in two ways by using a unique dataset obtained from an experiment where all the packets crossing a backbone router are captured. First, this enables an examination of the validity of the fundamental assumptions underlying the model across several links, covering a large range of bandwidths and utilization levels. Second, we extend the model from links to a network node, by examining the merging and splitting properties of the (sub)streams through the router, and mapping these to the merging and splitting properties of the model. We show how the model can, in most cases, capture the observed multiplexing and demultiplexing behaviour of the router, opening up the possibility of its use for understanding traffic flows in networks. We show that failures in the model cannot be accounted for simply through considering utilisation levels, and explain how they can in fact be used as a detector of upstream bottlenecks and traffic shaping.

Traffic shaping function becomes imperative for the new broadband services that are being deployed in order to avoid information loss, to provide the end users multiple traffic options in terms of bandwidth and to ensure optimal use of the communication channels. To simultaneously manage the amount of cell loss and delay experienced by two or more classes of service categories constant bit rate/ variable bit rate, we developed a new buffer partitioning scheme tagged complete sharing with gradual release. The proposed model was combined with a scheduling method known as weighted round robin with absolute increment. An analytical model was developed for the proposed buffer partition to dynamically monitor and determine the output mean rate of the classes of service present, and the individual mean rate of the class of service. The model was simulated and performance evaluation carried out. The result thus obtained depicts a better performance as a method of traffic shaper in a multi-quality of service traffic over asynchronous transfer mode networks.

Belakangan ini banyak ISP yang memberikan layanan broadband Internet dengan bandwidth yang sangat besar mulai dari 512Kbps hingga 2Mbps. Umumnya layanan ini digunakan oleh Warung Internet dan perusahaan-perusahaan yang bergerak dibidang Internet. Sekarang tinggal bagaimana mengatur bandwidth sebesar itu agar dapat dimanfaatkan bersama secara " adil ". praktikum ini akan membahas mengenai pembuatan sebuah bandwidth manager yang secara transparan dapat diletakkan pada jaringan yang telah ada. Pengenalan QoS Mungkin Anda semua sudah mengetahuinya, QoS adalah singkatan dari Quality of Service. Tetapi apa artinya? Apa yang bisa Anda lakukan untuk mengimplementasikannya? QoS artinya network yang telah memenuhi kriteria-kriteria tertentu yang dibuat oleh manager-manager network tersebut. Kriteria tersebut termasuk: 1. Availability, yaitu persentase hidupnya sistem atau layanan yang diberikan. Idealnya, availability harus mencapai 100% atau setidaknya 99,9999% (ada 6 buah angka sembilan), yang menunjukkan tingkat ke rusakan sebesar 2,6 detik per bulan. 2. Bandwidth Usages, yaitu masing-masing user akan diberikan bandwidth sesuai kebutuhan mereka. Selain berdasarkan user, pemakaian bandwidth juga dapat dibagi berdasarkan jenis traffic atau jenis protocol. Misalnya, untuk traffic HTTP dialokasikan bandwidth sebesar 40%, untuk voice 25%, sisanya untuk ftp dan lain-lain. Yang terakhir ini sering disebut dengan " traffic shaper ". 3. Throughput, yaitu kecepatan (rate) transfer data efektif, yang diukur dalam bps (bit per second). Penggunaan sebuah saluran secara bersama-sama akan mengurangi nilai ini. 4. Latency, adalah waktu yang dibutuhkan data untuk menempuh jarak dari asal ke tujuan. Tundaan (delay) ini dapat dipengaruhi oleh jarak (misalnya akibat penggunaan wireless LAN), atau kongesti (yang memperpanjang antrian), atau bisa juga akibat waktu olah yang lama (misalnya proses baca-tulis pada sebuah proxy server). 5. Packet Loss, yaitu jumlah paket yang hilang. Umumnya perangkat network memiliki buffer untuk menampung data yang diterima. Jika terjadi kongesti yang cukup lama, buffer akan penuh, dan data baru tidak dapat diterima. Paket yang hilang ini harus diretransmisi, yang akan membutuhkan waktu tambahan. Ada banyak kriteria-kriteria lain yang dapat ditentukan oleh seorang manager network disebuah perusahaan Lebih lengkap mengenai QoS dapat dilihat di http://www.qosforum.com/ tech_resources.htm Pada kasus tertentu, sebuah perusahaan bahkan memilih ISP yang telah memiliki service-level agreements, atau SLA. Dengan adanya perjanjian ini, perusahaan tersebut dijamin akan mendapatkan layanan sesuai yang diiklankan oleh ISPnya atau akan mendapat kompensasi apabila layanan tersebut tidak sesuai yang diharapkan. Praktikum ini secara khusus akan membahas bagaimana sebuah akses Internet berkecepatan tinggi (broadband Internet access) dapat dimanfaatkan bersamaan oleh beberapa karyawan di sebuah perusahaan. Dengan menggunakan sebuah fasilitas bandwidth manager, masing-masing karyawan akan dialokasikan dengan bandwidth sesuai kebutuhan mereka.

This survey looks at how traditional networking techniques (e.g., caching, traffic shaping, path diversity, and load balancing) have been adapted to address the needs of Internet-based video delivery. The stringent timing and relatively high bandwidth requirements of video traffic are taxing on best-effort networks and many video specific protocols and delivery methods have emerged over time in an attempt to mitigate network limitations. Video quality is directly tied to the underlying networks' ability to deliver data in time for playout. This paper surveys three classes of techniques which have been proposed for improving the quality of Internet delivered video: network load reduction, network interruption mitigation, and network load distribution. We discuss how each of these paradigms is applied within the different segments of the end-to-end video delivery system: by the server, in the network, or at the client, with a focus on how the underlying network conditions affect video quality optimization.

The issue of what discriminatory use of a network means has arisen in two recent decisions of the United States and Canadian federal communications commissions, the FCC and the CRTC respectively. The topic is a contemporary and hotly debated one, as when a course is fixed it will strongly influence the future of the Internet. It can be stated as the dichotomy of open and competitive or closed and oligopolistic. A study and comparison of the two different approaches is vital to clarify the debate, and hopefully guide Canadian policy in a direction that will benefit the whole community.

Abstract—Many Internet events exhibit periodical patterns. Such events include the availability of end-hosts, usage of internetwork links for balancing load and cost of transit, traffic shaping during peak hours, etc. Internet monitoring systems that collect huge amount of data can leverage periodicity information for improving resource utilization. However, automatic periodicity inference is a non trivial task, especially when facing measurement “noise”.

Multimedia transmission over communication networks has been growing rapidly with respect to the number of users and the amount and type of traffic. Wireless communications are a fast emerging technology and have become an essential feature of everyday life. Not only are computer networks becoming mobile, but each device is eventually having one or several wireless interfaces (e.g. laptops, cameras, phones etc.).Simultaneously, multimedia is having an equivalent growth, and is now constituting a large part of the traffic flowing through the wireless network through increased use in both the commercial as well as the social arena. These multimedia applications such as voice over IP, video conferencing, and video-on-demand, generate traffic with characteristics that differ significantly from the traffic generated by data applications, and they have more strict delay and packet loss requirements. Guaranteeing those requirements in wireless environments is very challenging because wireless links have variable characteristics especially through their mobile and ad hoc nature. To deal with this problem, many wireless communication standards have been defined enhancing the Quality of Service (QoS) of the whole system but do not specifically enhance multimedia traffic on its own. In this paper we propose mechanisms for modifying the existing QoS frameworks to include traffic prioritization, with multimedia traffic as the highest priority and to include different dropping probabilities for each class and congestion avoidance. We first study existing QoS frameworks then modify an existing framework (ASAP) to include traffic classification and bandwidth adaptation. iii ACKNOWLEDGEMENTS First and foremost I would like to extend my sincere appreciation to my supervisor Mr B Nyambo for his unwavering support, tolerance and advice, he introduced me to NS2 and MANETS so this research study would not have been a success without his help. Secondly I would like to thank my mother Ireen Nyarota for raising me up and still taking care of me even though Iam now an old man. I would also like to thank Tendai Muza for the love,the whole Computer Science department for welcoming me into this wonderful computer science family. All my friends , students and everyone who helped me in my masters degree. Above all I thank Jesus Christ my savior, the holy spirit my guardian, and Jehovah the father without you I am nothing at all.

Adaptive video streaming is based on the concept of allowing the quality level of a video stream to change during its lifetime based on certain parameters. This approach makes the video stream more tolerant with respect to fluctuations in available bandwidth during the session lifetime. In order for a video stream to decide which quality level to request from the server, it is typical to use both parameters from the client side such as CPU load and estimated parameters from the network side such as available bandwidth. In order for available bandwidth estimations to become more accurate, it is beneficial with some degree of traffic shaping. This paper describes a new method of achieving a traffic shaping effect for adaptive video streaming sessions delivered to members of a specific user group identified by their unique network destination. The novelty of the method is represented by the use of segment size control per video session, rather than active shaping of each session. The effect of the method is analyzed by means of measuring packet IAT and video segment fetch durations, in a controlled lab environment using the Smooth Streaming framework from Microsoft.

This survey looks at how traditional networking techniques (e.g., caching, traffic shaping, path diversity, and load balancing) have been adapted to address the needs of Internet-based video delivery. The stringent timing and relatively high bandwidth requirements of video traffic are taxing on best-effort networks and many video specific protocols and delivery methods have emerged over time in an attempt to mitigate network limitations. Video quality is directly tied to the underlying networks' ability to deliver data in time for playout. This paper surveys three classes of techniques which have been proposed for improving the quality of Internet delivered video: network load reduction, network interruption mitigation, and network load distribution. We discuss how each of these paradigms is applied within the different segments of the end-to-end video delivery system: by the server, in the network, or at the client, with a focus on how the underlying network conditions affect video quality optimization.

In Enterprise Data Centers (EDC), service providers are usually governed by Client Service Contracts (CSC) that specify, among other requirements, the rate at which a service should be accessed. The contract limits the rate to no more than a number of service requests during a given observation period. In two-tier setups, a cluster of Service-Oriented Networking (SON) Appliances form a pre-processing tier that accesses services in the service tier. SON Appliances locally shape the flow of requests to enforce the global rate defined in the CSC. Off-the-shelf SON Appliances present architectural limitations that prevent them from being used to efficiently perform traffic shaping in the presence of multiple service hosts. In this paper, besides identifying these limitations, we provide two contributions in this field. First, we introduce a SON Appliance architecture fit for multi-service traffic shaping. Second, we propose and validate an algorithm for multipoint-to-multipoint service traffic shaping in two-tier EDCs. We show via simulation that our approach solves the multipoint-to-multipoint service traffic shaping problem while pushing the system to its maximum capacity.

Traffic shapers are used by researchers to emulate the behavior of networks and applications in test environments, typically with user-defined traffic shaping parameters such as throughput and loss. Also, traffic shapers are used for the enforcement of SLA, so they are of interest for Internet Service Providers. However, output given by traffic shapers may not be as accurate as desired. Therefore, it is important to assess the accuracy of the implementation of traffic shapers. In this paper, we evaluate two traffic shapers with regard to the performance of their throughput shaping. For this evaluation, traces were collected. The properties of the resulting throughput at the outlet of the shaper are compared to the properties of the throughput at the inlet in combination with the preset shaper parameters. In this sense, we also compare shapers installed on Advance Micro Devices (AMD) and Intel platforms, and we use different PDU sizes and load levels to test the influence of those parameters on the shaping. We are furthermore able to deduct internal shaper parameters such as packet rate and buffer size from our measurements, and we analyse the statistical properties of the packet departure process. The extensive measurement results in this paper allow for a detailed assessment of the question whether the shaper performance is up to mark for a desired timescale. In general, the performance of both shapers and hardware platforms can be considered satisfactory on the investigated time scales between 1 ms and 1 s, with a slight advantage for NetEm on AMD.

The impact of buffer management and priority scheduling is examined in stressful scenarios when the aggregate incoming traffic is higher than the output link capacity of an Asynchronous Transfer Mode (ATM) traffic shaper. To simultaneously reduce cell loss and extreme delay behavior for two or more classes of service, we show that a dynamic priority scheme is required. We propose a scheduling algorithm where the priority of different service queues is dynamically modified to allow for the provisioning of isochronous services on one of the queues. Buffer management ensures that all service queues are guaranteed a minimum amount of memory, yet available memory can be shared between service queues when necessary. This approach guarantees that no cells are lost under strain conditions until all buffer is exhausted.

More than a decade ago it was discovered that some it with other estimators, two different types of simulation LAN traffic exhibits self-similar rather than Poisson behaviour studies were performed. The first one is using fractional and there is ongoing research towards finding and improving Gaussian noise (fGn) sequences generated by Dietrichsuitable estimators which may help to characterize various Newsam algorithm [8, 9], which generates exact self-similar types of network traffic. Such characterization can be potensequences. The second one is using a fractional tially applied for control purposes such as traffic shaping, load auegressivem ong one 1S usmg a II]. balancing, etc. The Hurst exponent is used to measure the in-autoregressive moving average (FARIMA) process [10, 11]. tensity of long-range dependence (LRD) in the network traffic. After that the same set of estimators is tested on the trace Despite having several existing estimators, LRD analysis is still files for MPEG-encoded musical videos and movies. impeded by the difficulty of actual identification of its intensity. The rest of the paper is organised as follows. Section II This paper continues work on estimating the Hurst exponent shows the relationship between the autocorrelation function from the autocorrelation function, a so-called HEAF estimator (ACF) and LRD. Section III introduces the HEAF estimator. introduced earlier by the authors. It also compares HEAF with a few existing estimators such as Wavelet, Higuchi, aggregated The results of simulation experiments are presented in secvariance time (V/T) and Rescaled-range (R/S). The simulation tion IV and experiments on MPEG-encoded movie traces in studies show that HEAF can be used to capture the LRD in the Section V. The properties of the HEAF estimator are shown network traffic if true process is fGn and FARIMA.

The ATh4 customer premises equipment (CPE) architecture proposed in this paper supports a broad range of user applications with varying service requirements. A backplane bus was chosen as a cost effective, reliable means of providing the necessary internal interconnection bandwidth. Incorporating a novel access mechanism and graceful performance degradation under fault conditions, this bus provides a foundation for Quality of Service provision. Essential functions including traffic management, connection admission control and signalling are then built on top of this. A prototype system which uses a semi-custom ASIC to implement the backplane access mechanism is briefly described. The architecture provides a flexible basis for service provision, allowing a single CPE device to meet a variety of end user requirements. Results of simulations showing the effectiveness of internal bandwidth access mechanisms are presented.

Networks-on-chip (NoC) for general-purpose multiprocessors require quality of service mechanisms to allow realtime streaming applications to be executed along with latency-sensitive general purpose processing tasks. In this paper, we propose a NoC link arbitration technique that supports bandwidth guarantees along with best effort latency optimizations. In contrast to many existing quality of service mechanisms, our technique prioritizes best effort over

According to recently presented QoS architecture by 3GPP [1], a traffic conditioner may be deployed to provide conformance of the negotiated QoS in UMTS. The traffic conditioning is performed by traffic shaping or/and policing. A framework of traffic conditioning with QoS provisioning in 3G radio access network is proposed in this paper. The main idea of our traffic conditioning approach is to employ traffic shaping at each UE and traffic policing at the RNC. The traffic generated at each UE is regulated by a traffic shaper in the form of a token bucket, and the conformance of the traffic is policed at the RNC according to traffic policing policies. A system model based on the proposed framework is implemented. The simulation results regarding the impact of traffic shaping on packet discarding probability, the tradeoff between probability of non-compliance and shaping delay are also presented in this paper.

Abstract: Using workload shaping technology, we present an approach to remove hardware over-provisioning implementing task buffers and scheduler, in terms of energy consumption. Task buffers reorder tasks with various priorities and routes them to appropriate virtual machines. Scheduler monitors the task buffering and hardware load status, and decides the optimal number of active physical and virtual machines. In addition, we designed a mechanism wherein tasks with fast executing are routed in fast and high energy consumption machines and slow tasks to slow and low energy consumption machines. As a result, our approach efficiently can shape workloads and manage the optimal number of active virtual machines and physical machines, in terms of energy consumption. To evaluate our approach, we generated synthetic workload data and evaluated it both in simulating and actual cloud environment. Our experimental results demonstrate our approach outperforms in terms of energy consumption to w...

We propose a minimum cost method for traffic shaping in the context of QoS-based networks. Given the user's desired QoS and the network's resource availability, our procedure determines the least-cost parameters for a traffic shaper which still guarantees access to the network whilst satisfying the QoS constraints. We illustrate our scheme using on-off sources and formulate the QoS constraints by effective bandwidths.

In this work, we propose the use of an Ant Colony Optimization (ACO) algorithm to mitigate packet loss in an optical packet switching network that carries selfsimilar traffic, which is known to have a great impact in the buffer performance in terms of loss probability as exemplified in this work. By adaptively routing the packets and balancing the network load, we demonstrate by some simulations the effectiveness of this approach when compared with a shortest-path routing scheme, achieving a performance that is comparable to the Poisson traffic scenario in some cases. The proposed algorithm can be used as a viable alternative to traffic shaping techniques.

Maximizing bandwidth utilization and providing performance guarantees, in the context of multimedia networking, are two incompatible goals. Heterogeneity of the multimedia sources calls for e ective tra c control schemes to satisfy their diverse Quality of Service(QoS) requirements. These include admission control at connection set up, tra c control at the source ends and e cient scheduling schemes at the switches. The emphasis in this paper is on tra c control at the source end. Most multimedia sources are bursty in nature. Tra c shapers have been mainly studied hitherto from the point of view of their e ectiveness in smoothing the burstiness. Leaky Bucket(LB) scheme, to cite an example, is a mean rate policer smoothing at the token generation rate. Studies on bursty sources show that burstiness promotes statistical multiplexing at the cost of possible congestion. Smoothing, on the other hand, helps in providing guarantees at the cost of utilization. Thus need for a exible scheme which can provide a reasonable compromise between utilization and performance is imminent. Recent studies 10, 12] have also questioned the suitability of LB for policing real-time tra c due to the excessive delays. We argue for a policy which is less stringent on short term burstiness than the LB. We propose a new tra c shaper which can adjust the burstiness of the input tra c to obtain reasonable bandwidth utilization while maintaining statistical service guarantees. The performance study is conducted in two parts. In the rst part, we study the e ect of varying the shaper parameters on the input characteristics. In the second part, we dimension our scheme and a LB equivalently and compare the mean and peak rate policing behavior with delay and loss as the performance parameters. Adopting a less stringent attitude towards short term burstiness is shown to result in considerable advantage while policing real-time tra c. Future research possibilities in this topic are explored.

A high capacity network adapter design is presented, using a CPU-based approach for shaping traffic, scheduling cells for multiplexing and controlling a DMA engine for a large number of ATM connections. By performing a scheduling operation for each cell, traffic shaping and data movement can be combined. Two scheduling algorithms are proposed, implemented and evaluated using detailed instrumentation on current high performance processors. We focus on the outbound direction since this is where most data is moved in a large network server, such as a high capacity WWW server. It is anticipated that the rapid increase in processing power can enable future servers to perform such scheduling in real time, even on very high capacity network connections. We present a general architecture for the adapter, discuss how shaping and scheduling can be combined, and present measurements that show how the proposed scheduling algorithms perform close to the real time requirements of high capacity network connections, even using current general-purpose microprocessors. We have chosen ATM for our evaluation, although the scheduling algorithms are general and may be employed in any network technology with flow reservations, such as the next generation Internet protocols.

The unreliable network layer in the Internet has been a cause of concern for network managers and users. The unreliability manifests itself in terms of lost or delayed packets due to congestion or failure in the network. One of the causes of congestion is the inherent burstiness of network traffic. Traffic includes unpredictable large file or web page transfers as

We propose to provide QoS differentiation at the IP level based on the DiffServ model over the standard Distributed Coordination Function (DCF) method of 802.11. By constraining the output rate of each host in a cell via DiffServ scheduling and traffic shaping mechanisms, we can keep the 802.11 network in the state of non-saturation so that the time-sensitive Expedited Forwarding (EF) class benefits from stable short delays. We analyze the performance of 802.11b to find how we should constrain the rate of DiffServ sources and present extensive measurements that show how the QoS differentiation is achieved. 0-7803-7975-6/03/$17.00 (C) 2003

Resumen--El crecimiento de las aplicaciones de comercio electrónico basado en Internet y los ataques a las redes se han vuelto más comunes y sofisticados. Las redes demandan medidas de protección más elaboradas para garantizar una segura operación y dar continuidad a los servicios críticos. Estas medidas incluyen métodos de detección y repuesta a los intentos de intrusión en tiempo real. En este trabajo se propone modelos estadísticos y clasificadores multivariables para detectar perfiles de tráfico anómalos y usar clasificadores basados en redes neuronales. Se muestran resultados numéricos para casos simples de ataques de denegación de servicio.

this paper presents the development of a new scheme called Adaptive Throughput Policing and Shaping (ATPS) algorithm to control internet inbound throughput and burst flow in a Campus IP-based network. Real live throughput collected from a Campus Network with Committed Access Rate (CAR) of 16 Mbps bandwidth speed are simulated and analyzed. New mathematical model with identified parameters on ATPS is derived. Adaptive throughput policies with burst shaping are simulated using Token Bucket theory control mechanism with 16 Mbps threshold policy. Three main adaptive policy conditions called P1, P2 and P3 which is controlled on 110%, 100% and 50% threshold rate are defined as filtered condition. Burst throughputs are shaped into next free bucket capacity for the next flow time. The throughputs are continuously shaped if the next bucket is full until free buckets are available. This new ATPS algorithm is numerically evaluated and analyzed on traffic performance which controlled the bandwidth and burst throughput. Performance results present reduced bucket capacity, reduced bandwidth rate in throughput transfer, no burst throughput and no byte loss in conforming traffic transferred in a network compared to previous implemented ATP algorithm, which held burst throughput and byte loss in the system.

Optical burst switching (OBS) has been proposed in the late 1990s as a novel photonic network architecture directed towards efficient transport of IP traffic. OBS aims at cost-efficient and dynamic provisioning of subwavelength granularity by optimally combining electronics and optics. Optical bursts cut through intermediate nodes, i. e., data stays in the optical domain at all times, while the control information is signaled out-of-band and processed electronically. In contrast to optical packet switching, OBS aggregates and assembles packets electronically into bursts of variable length according to destination and QoS class at the edge of the network.