Observingtheeffectof TCP congestion controlon networktraffic

Save this PDF as:
 WORD  PNG  TXT  JPG

Size: px
Start display at page:

Download "Observingtheeffectof TCP congestion controlon networktraffic"

Transcription

1 Observingtheeffectof TCP congestion controlon networktraffic YongminChoi 1 andjohna.silvester ElectricalEngineering-SystemsDept. UniversityofSouthernCalifornia LosAngeles,CA Abstract In this paper we quantitatively observe the relationship between packet loss rate and degree of traffic burstiness using ns-2 network simulations. We examine the traffic behavior of a single TCP connection in a lossy environment. The loss agent in the simulation uniformly drops incoming packets with a given probability. By changing the packet loss rate, we can vary the congestion level of network and influence the behavior of TCP s congestion control mechanism. We measure the variation of traffic burstiness as the packet loss rate is changed. We also observe different traffic behavior with respect to the types of traffic sources. It is observed that for exponential traffic source, the degree of traffic burstiness is influenced by the packet loss rate over some range. However, the degree of traffic burstiness for Pareto traffic source is almost the same regardless of the packet loss rate. While our simulation is similar to previous works [1][2], the observation is interesting since the relationship between packet loss rate and degree of traffic burstiness is measured quantitatively. Index Terms TCP congestion control, exponential back-off, heavy-tailed distribution, traffic characterization and modeling. I. INTRODUCTION Self-similarity is a well-known phenomenon in a number of measurements over a wide range of networks including Ethernet, wide area network, and WWW (world wide web). In networking area, self-similarity means qualitatively that when we aggregate a traffic process with respect to timescale, correlation structure of the aggregated process is similar to that of the original traffic process. In other words, bursty traffic remains bursty when we look at the traffic at large timescales. Self-similarity manifests itself mathematically in a number of equivalent ways. One example of those representations is long-range dependence in which the autocorrelation of traffic decays more slowly than that of a traditional Markovian (or memoryless) process does. While the autocorrelation function of a memoryless process decays exponentially, the autocorrelation function of a self-similar process decays hyperbolically fast and it implies strong correlation structure in such a process. We refer to [3][4][5] for exact definition and properties of self-similar processes. There have been diverse research efforts to investigate various aspects of traffic self-similarity, for example, modeling techniques, impact on network performance, and explanations 1 Yongmin Choi is a Ph.D. student at the University of Southern California. He is now with Service Development Laboratory at KT Corporation, Seoul , Korea. for the presence of self-similar traffic. To explain why selfsimilar traffic arises, many researchers have examined the traffic in two different perspectives: statistical properties of traffic source (especially with heavy-tailed distributions) and networking mechanisms (TCP flow/congestion control). It is shown that file sizes, flow (or session) durations with heavy-tailed distributions can induce self-similar traffic [6][7][8]. This perspective mainly deals with application/session layer quantities such as file size and session durations, or with user behaviors such as user think time. Regarding the networking mechanisms, we reviewed the following two works. In [1], the authors claim that the TCP congestion control mechanism generates heavy-tailed off periods in the traffic transmission pattern, which introduces long-range dependence in the traffic. They argue that the exponential backoff mechanism extends the packet inter-departure time distribution and that the traffic shows pseudo self-similarity in that scaling behaviors of traffic exist only over a finite range of timescales. In another work [2], the authors assert that not only the exponential back-off but the congestion avoidance mechanism is also responsible for the traffic self-similarity. They argue that at low packet loss probability, a sustained correlation appears due to congestion avoidance and at high probability the correlation appears due to the timeout mechanism. In this paper, we extend the previous research findings and observe effect of TCP congestion control mechanisms upon the degree of self-similarity in data network traffic. In our simulation, we observe traffic behavior of a single TCP connection in lossy environment. As we vary the packet loss rate of a TCP connection (i.e. network congestion level), TCP adaptively changes its congestion window size and packet transmission rate to cope with available bandwidth. The packet interdeparture time distribution for the resulting TCP traffic is affected by the packet loss rate and the degree of traffic burstiness varies in proportion as the distribution is changed. We carefully observe the relationship between packet loss rate and degree of self-similarity in traffic using ns-2 network simulation. Specifically, we observe the influence of TCP congestion control on the packet inter-arrival time distribution and the degree of traffic burstiness. For the traffic source, we use two types of packet inter-arrival distributions: exponential and Pareto. It is shown that the packet inter-departure time distribution at the TCP sender node is the same as the original inter-arrival distribution when the packet loss rate is low. How Applied Telecommunications Symposium

2 ever, when the packet loss rate is high, the distribution becomes heavy-tailed and looks similar regardless of the type of traffic source. On the other hand, the degree of traffic burstiness depends on the type of traffic source. For the exponential source, the Hurst parameter is changed as the packet loss rate is increased. However, there is no apparent change in the Hurst parameter for the Pareto source over the range of packet loss rate we experimented. While our simulation is similar to the works in [1] [2], the relationship between packet loss rate (i.e. heavy-tailedness of packet inter-departure time distribution) and degree of selfsimilarity is examined quantitatively. The rest of this paper is organized as follows. In section 2, we briefly summarize the related mathematical concepts in selfsimilar traffic. Section 3 presents our simulation setup and scenario. Section 4 provides simulation results and discussions on the results. We summarize future research direction in section 5. II. BACKGROUNDS In this section, we briefly review some related concepts in self-similar traffic analysis. A more formal description can be found in [9] and references therein. A real-valued, discrete time random process X = {X t : t = 0, 1, 2,...}, is called a covariance stationary process if it has a finite mean µ, a finite variance R(0), and a time-homogeneous covariance function. The time-homogeneous covariance function R(k) is defined as R(k) = E{(X t µ)(x t+k µ)}, k = 0, ±1, ±2,... (1) where µ = E[X t ] < and r(0) <. For each m = 1, 2,..., we define the m-aggregated process X (m) = {X (m) t, t = 0, 1, 2,...} by summing the original time series over nonoverlapping, adjacent blocks of size m as follows: X (m) t = 1 m tm i=(t 1)m+1 X i (2) A discrete-time process X t is said to be exactly self-similar if for all m = 1, 2,..., the autocorrelation functions of the m- aggregated process and the original process are the same, i.e., R X (m)(k) = R X (k) (3) There is a weaker condition of self-similarity. A process is said to be asymptotically self-similar if the autocorrelation function of m-aggregated process is asymptotically the same as that of the original process for all m large enough R X (m)(k) R X (k) as m (4) We can interpret the aggregation of time series as compression of the timescale so that the original process X t has the highest resolution possible and the m-aggregated process is the same process reduced in resolution by a factor of m [10]. By averaging over each set of m points, we lose the fine details of the original process available at the highest resolution. With this definition of self-similarity, the autocorrelation of the aggregated process has the same form as that of the original process. This implies that the degree of variability, or burstiness in self-similar traffic would be the same at different timescales. Mathematically self-similarity is observed in a number of equivalent ways: slowly decaying variance, long-range dependence, and power-low behavior near the origin (1/f-noise) [3]. For a self-similar process, the variance of the time average does not go to zero as quickly as that of an ergodic process. The time average of an ergodic process should equal an ensemble average and the variance of the time average goes to zero relatively quickly as m becomes large. On the other hand, the variance of the time average for a self-similar process does so more slowly (slowly decaying variance). Thus, we have a self-similar process X with parameter β (0 < β < 1) if it satisfies for all m = ±1, ±2,..., V ar(x (m) ) = V ar(x) m β (5) An ergodic process has a parameter β = 1. For a self-similar process, the degree of burstiness (or selfsimilarity) is represented by a single parameter called the Hurst parameter, H. The Hurst parameter is related with the slope parameter β as follows: H = 1 β/2 (6) To measure the degree of burstiness for the synthetic traffic generated by the simulation, several estimation methods are used. In our simulation, we extensively use the wavelet-based tool [11]. The tool estimates the variance of wavelet coefficients of the packet series at particular timescales. This estimate is then plotted in a log-log scale diagram and the least square fit over all timescales is calculated. Then the slope of this asymptotic linear region gives an estimate of the Hurst parameter. Before performing the wavelet analysis, the synthetic traffic trace was aggregated into bins with size smaller than one RTT. The aggregated time series was analyzed using the publicly available tools. Lastly we review heavy-tailed distribution which is related to the self-similar traffic. As explained in the introduction, the traffic self-similarity can attributed to the application or session layer quantities with heavy-tailed distributions such as file sizes, flow (or session) durations [6][7][8]. Heavy-tailed distributions are introduced to explain the traffic self-similarity with TCP flow/congestion control [1][2]. The packet interarrival time distribution is classified as heavy-tailed over the finite range of timescales. A distribution is defined as heavytailed if the asymptotic behavior of the distribution follows the power-law with exponent less than 2, i.e., P [X x] x α, as x, 0 < α < 2 (7) When the packet inter-arrival time distribution is heavy-tailed, the resulting traffic is self-similar with the Hurst parameter given as follows: H = 3 α (8) Applied Telecommunications Symposium

3 Incoming traffic ( or Pareto) S Loss Agent R Complementary CDF of packet interarrival time Pareto (α=1.5) Fig. 1. Simulation setup III. SIMULATION SCENARIO AND SETUP This section describes our simulation setup and scenario. We use a simple experiment with the ns-2 simulator to observe the relationship between packet loss rate and traffic burstiness. As in other similar studies, we simulate the case in which TCP connections compete to send packets through a bottleneck link. Since we are interested in the behavior of a single TCP flow, we simulate only a single connection transmitting over a lossy link instead of simulating many TCP flows competing for the bottleneck resource. The interaction among competing TCP flows is modeled into a loss agent that uniformly drops packets with a given probability. The simulation topology is shown in Fig. 1. In this figure, node S is the sender, node R is the receiver. We put a loss module into the link connecting from node S to node R. Thus, only packets from the sender to the receivers experience losses. The packet size is fixed at 1000 bytes. The link capacity, the buffer size at the sender node, and the receiver window size are set to be large so that only the loss module affects TCP performance. The two-way propagation delay between the source S and the receiver R is set to 40 msec. TABLE I SIMULATION PARAMETERS Parameter Value Link bandwidth 128 kbps Link delay 20 msec Buffer size 100 Avg. inter-arrival time 1/10 seconds Total simulation time seconds The sender objects acts as an infinite source; thus it always wants to send as much data as possible. For the traffic source, we consider two different packet inter-arrival time distributions: exponential and Pareto. The packet arrivals in the exponential traffic source are independent while those arrivals in the Pareto traffic source are strongly correlated. We compare the degree of self-similarity to confirm how TCP modifies traffic self-similarity with respect to the source characteristics. We trace packet arrival events only at the downstream link of node D. IV. RESULTS The goal of this simulation study is to observe the influence of TCP congestion control upon the network traffic. The application generated traffic at the sender node is modulated by TCP flow control and thus the outgoing traffic observed at the link shows different statistics from the incoming traffic. Furthermore, packet losses at the link add more variability to TCP congestion control. We adjust the congestion level of the link Fig Complimentary CDF of packet inter-arrival time for incoming traffic by changing the packet loss rate and observe the statistics of outgoing traffic at the link connecting the sender and receiver nodes. To measure the burstiness of TCP traffic, we will use such tools as histogram and complementary CDF of packet inter-departure time, coefficient of variation (c.o.v.), and degree of self-similarity for the traffic. We also find a linear relationship between packet loss rate and degree of self-similarity. A. Probabilistic distribution of incoming traffic In this simulation, we use two packet inter-arrival time distributions for the incoming traffic source at the TCP sender: exponential and Pareto distributions. The complementary distributions of two traffic sources are shown in Fig. 2. For the exponential source, the packet arrival rate is λ=10[packets/sec]. The parameters of Pareto traffic source is adjusted to have the same mean arrival rate as the exponential source has. For Pareto distribution, the mean arrival rate is given as: E[X] = α α 1 x 0 (9) Since we choose the shape parameter α=1.5 for the Pareto distribution, the minimum value x 0 of Pareto distribution should be 1/30 in order to give the same mean arrival rate as the exponential source. B. Packet inter-departure time distribution The inter-arrival time distribution of incoming packets is changed by TCP flow control. We observe the inter-departure time distribution of packets after the TCP sender node for both traffic sources. In addition to TCP flow control, packet losses at a link can change TCP status. When the network is highly congested, the exponential back-off mechanism of TCP works to relieve the network congestion. At this mode, packets are transmitted at the multiples of RTT estimate so that the packet inter-departure time can be increased up to 64 times the initial RTT estimate value. We observe this modification effect of TCP congestion control prominently at high packet loss rate. The complementary cumulative distribution functions for the original packet inter-arrival time and the inter-departure time of TCP traffic at high packet loss cases (p = 0.2) are shown in Fig. 3 for Applied Telecommunications Symposium

4 Complementary CDF of packet interarrival time (RTO=6 sec) Complementary CDF of packet interarrival time p=0 p=0.001 p=0.01 p=0.1 p= Complementary CDF of packet interarrival time (RTO=6 sec) Pareto Complementary CDF of packet interarrival time Pareto p=0 p=0.001 p=0.01 p=0.1 p= Fig. 3. Complementary CDFs of packet inter-departure time at the packet loss rate (p = 0.2) (a) exponential (b) Pareto. Fig. 4. Complimentary CDF of packet inter-arrival time for TCP traffic (Up) (Down) Pareto both traffic sources. In this figure, we see the modulation effect of TCP on the incoming traffic such that the packet inter-arrival time distribution is changed at the high packet loss rate. We also identify that the packet inter-departure time distributions are similar regardless of traffic sources. Thus we can conclude that the packet inter-departure time distribution is mainly determined by the TCP congestion control especially at the high packet loss rate. Next, we draw the complementary CDF s for both sources by changing the packet loss rate in Fig. 4. In this figure, we see that the packet inter-departure time distribution is similar to that of the original source at low packet loss rates. As the packet loss rate is increased, the complementary CDF of packet interdeparture time is changed from the original form. The CDF becomes heavy-tailed rather than exponential for the exponential source. For the Pareto source, the CDF is still heavy-tailed but the slope is changed a little. The more interesting phenomenon is that the complementary distributions for both traffic source are similar to each other at high packet loss rates (p= 0.1 and 0.2). This result is also confirmed from the Hurst parameter and the coefficient of variation. Next, we examine the characteristics of TCP modification effect with the histogram of packet inter-departure time. The histogram of packet inter-departure time for TCP traffic is characterized by some dominant values. This phenomenon is rather obvious from the TCP congestion control mechanism. When the network is highly congested, TCP congestion control is working so that the packet inter-departure time should be multiples of retransmission timeout value (RTO). Thus, the packet inter-departure time is not arbitrary and its distribution has some dominant values (Fig. 5). C. Degree of traffic burstiness We investigate the degree of burstiness in the TCP traffic with two measures, the coefficient of variation and the Hurst parameter. Firstly we consider the coefficient of variation (c.o.v.) for measuring the modulation effect of TCP congestion control. The c.o.v. is defined as the ratio of the standard deviation to the mean of the observed packet inter-departure time. The c.o.v. gives a normalized value for the spread of a distribution and allows for the comparison of spreads of packet inter-departure time distributions over varying packet loss rates. Since TCP traffic is self-similar over limited range of time scales (i.e. pseudo self-similar), we admit that the Hurst parameter alone is not appropriate to describe the burstiness of TCP traffic. For both traffic sources, the coefficients of variations are shown in II. The coefficient of variation for the exponential source is 1. For the exponential source, the c.o.v. for outgoing TCP traffic is decreased when the packet loss rate is relatively low. However, the c.o.v. is increased when the packet loss rate is high. We infer that TCP regulates the incoming traffic with little packet Applied Telecommunications Symposium

5 TABLE II COEFFICIENT OF VARIATION OF PACKET INTER-DEPARTURE TIME IN TCP TRAFFIC Pareto Histogram of packet interarrival time (RTO=6 sec) 0.95 Trend of Hurst parameter vs. Packet loss probability Number of frequency Hurst parameter Interarrival time [sec] Histogram of packet interarrival time (RTO=6 sec) Pareto Probability Fig. 6. Variation of the Hurst parameter with respect to the packet loss rate (0.01 < p < 0.1) for exponential source. Number of frequency Interarrival time [sec] Fig. 5. Histogram of packet inter-departure time (p = 0.2) (Up) (Down) Pareto. loss so that the packet inter-departure time is constant if there are enough packets at the sender node. If the packet loss rate is high, then the packet inter-departure time has more variability by the exponential back-off mechanism. Like the complementary distribution of packet inter-departure time, there is an abrupt change in the c.o.v. between the packet loss rates 0.01 and 0.1 for the exponential source. We also note that for the Pareto source, the c.o.v. does not change much as in the case of complementary distribution. However, we observe a small decrease in the c.o.v. at p = 0.1. Next, we observe the Hurst parameter as a measure of selfsimilarity in TCP traffic. Although the TCP traffic is not strictly self-similar, it is still assumed to be pseudo self-similar or longrange dependent over a limited range of time-scales in which non-degenerate correlation structure exists. Hence, we use the Hurst parameter as a measure of traffic self-similarity over the range of time-scales. As shown in Table 2, the change in the degree of traffic self-similarity (i.e. the Hurst parameter in our experiment) with respect to the packet loss rate is different for both traffic sources. For the Pareto source, the Hurst parameter does not change a lot except a small increase at p = 0.1. For the exponential source, the Hurst parameter changes little at low packet loss rates. However, the degree of self-similarity is increased as the packet loss rate exceeds 0.1 and the traffic shows self-similarity like that of the Pareto source. We conclude that TCP induced self-similarity is apparent at high packet loss rate (i.e. in a highly congested network). For the exponential source, we investigate in depth the transition in value of the Hurst parameter between packet loss rates 0.01 and 0.1. The variation of the Hurst parameter with respect to the packet loss rate is shown in Fig. 6. There exists a quasilinear relationship between the Hurst parameter and packet loss rate. V. CONCLUSION This paper quantitatively observes the relationship between packet loss rate and the degree of burstiness in traffic using ns- 2 network simulations. By changing the packet loss rate of a link, we adjust the congestion level of the link and influence the behavior of TCP traffic. We observe the distribution of TCP traffic and measure the degree of burstiness with the coefficient of variation and the Hurst parameter. Using two traffic sources, exponential and Pareto distribution, we quantitatively measure TCP induced burstiness in addition to source induced burstiness. In addition, we find that there exists a relationship between the packet loss rate and the Hurst parameter for the exponential traffic source. It is interesting to observe the relationship over the more extensive range of packet loss rate. To explain this modification effect of TCP, we use Markovian Applied Telecommunications Symposium

6 models for TCP congestion control. Extensive results and detailed explanation of the relationship will be reported in another paper. REFERENCES [1] L. Guo, M. Crovella, and I. Matta, How does TCP generate pseudo-selfsimilarity?, in Proc. of Ninth International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, 2001, pp [2] D. R. Figueiredo, B. Liu, V. Misra, and D. Towsley, On the autocorrelation structure of TCP traffic, Tech. Rep., Computer Science Department, University of Massachusetts, [3] W. E. Leland, M. S. Taqqu, W. Willinger, and D. V. Wilson, On the self-similar nature of Ethernet traffic (Extended Version), IEEE/ACM Trans. Networking, vol. 2, pp. 1 15, February [4] B. Tsybakov and N. D. Georganas, On self-similar traffic in ATM queues: Definitions, overflow probability bound, and cell delay distribution, IEEE/ACM Trans. Networking, vol. 5, pp , June [5] B. Tsybakov and N. D. Georganas, Self-similar processes in communications networks, IEEE Trans. Inform. Theory, vol. 44, pp , September [6] V. Paxson and S. Floyd, Wide-area traffic: The failure of poisson modeling, IEEE/ACM Trans. Networking, vol. 3, pp , June [7] M. Crovella and A. Bestavros, Self-similarity in world wide web traffic: Evidence and possible causes, IEEE/ACM Trans. Networking, vol. 5, pp , December [8] W. Willinger, M. S. Taqqu, R. Sherman, and D. V. Wilson, Selfsimilarity through high-variability: Statistical analysis of Ethernet LAN traffic at the source level, IEEE/ACM Trans. Networking, vol. 5, pp , February [9] R. J. Adler, R. E. Feldman, and M. S. Taqqu, A practical guide to heavy tails: Statistical techniques and Applications, Birkhauser, [10] W. Stallings, High-speed networks: TCP/IP and ATM design principles, Prentice-Hall, [11] D. Veitch and P. Abry, A wavelet based joint estimator of the parameters of long-range dependence, IEEE Trans. Inform. Theory, vol. 45, pp , April Applied Telecommunications Symposium

Examining Self-Similarity Network Traffic intervals

Examining Self-Similarity Network Traffic intervals Examining Self-Similarity Network Traffic intervals Hengky Susanto Byung-Guk Kim Computer Science Department University of Massachusetts at Lowell {hsusanto, kim}@cs.uml.edu Abstract Many studies have

More information

Internet Traffic Variability (Long Range Dependency Effects) Dheeraj Reddy CS8803 Fall 2003

Internet Traffic Variability (Long Range Dependency Effects) Dheeraj Reddy CS8803 Fall 2003 Internet Traffic Variability (Long Range Dependency Effects) Dheeraj Reddy CS8803 Fall 2003 Self-similarity and its evolution in Computer Network Measurements Prior models used Poisson-like models Origins

More information

Evaluation of Effective Bandwidth Schemes for Self-Similar Traffic

Evaluation of Effective Bandwidth Schemes for Self-Similar Traffic Proceedings of the 3th ITC Specialist Seminar on IP Measurement, Modeling and Management, Monterey, CA, September 2000, pp. 2--2-0 Evaluation of Effective Bandwidth Schemes for Self-Similar Traffic Stefan

More information

ON THE FRACTAL CHARACTERISTICS OF NETWORK TRAFFIC AND ITS UTILIZATION IN COVERT COMMUNICATIONS

ON THE FRACTAL CHARACTERISTICS OF NETWORK TRAFFIC AND ITS UTILIZATION IN COVERT COMMUNICATIONS ON THE FRACTAL CHARACTERISTICS OF NETWORK TRAFFIC AND ITS UTILIZATION IN COVERT COMMUNICATIONS Rashiq R. Marie Department of Computer Science email: R.R.Marie@lboro.ac.uk Helmut E. Bez Department of Computer

More information

Connection-level Analysis and Modeling of Network Traffic

Connection-level Analysis and Modeling of Network Traffic ACM SIGCOMM INTERNET MEASUREMENT WORKSHOP Connection-level Analysis and Modeling of Network Traffic Shriram Sarvotham, Rudolf Riedi, Richard Baraniuk Abstract Most network traffic analysis and modeling

More information

CoMPACT-Monitor: Change-of-Measure based Passive/Active Monitoring Weighted Active Sampling Scheme to Infer QoS

CoMPACT-Monitor: Change-of-Measure based Passive/Active Monitoring Weighted Active Sampling Scheme to Infer QoS CoMPACT-Monitor: Change-of-Measure based Passive/Active Monitoring Weighted Active Sampling Scheme to Infer QoS Masaki Aida, Keisuke Ishibashi, and Toshiyuki Kanazawa NTT Information Sharing Platform Laboratories,

More information

Self-Similarity Defined

Self-Similarity Defined Presentation Outline 1. Self similarity in nature 2. Quick review of autocorrelation 3. Definition of self-similar discrete process Exactly/asymptotic self-similar Long range vs short range dependence

More information

Defending Against Traffic Analysis Attacks with Link Padding for Bursty Traffics

Defending Against Traffic Analysis Attacks with Link Padding for Bursty Traffics Proceedings of the 4 IEEE United States Military Academy, West Point, NY - June Defending Against Traffic Analysis Attacks with Link Padding for Bursty Traffics Wei Yan, Student Member, IEEE, and Edwin

More information

Network traffic: Scaling

Network traffic: Scaling Network traffic: Scaling 1 Ways of representing a time series Timeseries Timeseries: information in time domain 2 Ways of representing a time series Timeseries FFT Timeseries: information in time domain

More information

Probabilistic properties and statistical analysis of network traffic models: research project

Probabilistic properties and statistical analysis of network traffic models: research project Probabilistic properties and statistical analysis of network traffic models: research project The problem: It is commonly accepted that teletraffic data exhibits self-similarity over a certain range of

More information

Measurement and Modelling of Internet Traffic at Access Networks

Measurement and Modelling of Internet Traffic at Access Networks Measurement and Modelling of Internet Traffic at Access Networks Johannes Färber, Stefan Bodamer, Joachim Charzinski 2 University of Stuttgart, Institute of Communication Networks and Computer Engineering,

More information

Connection-level Analysis and Modeling of Network Traffic

Connection-level Analysis and Modeling of Network Traffic Connection-level Analysis and Modeling of Network Traffic Shriram Sarvotham, Rudolf Riedi, Richard Baraniuk Abstract Most network traffic analysis and modeling studies lump all connections together into

More information

UNDERSTANDING the nature of network traffic is critical

UNDERSTANDING the nature of network traffic is critical IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 5, NO. 6, DECEMBER 1997 835 Self-Similarity in World Wide Web Traffic: Evidence and Possible Causes Mark E. Crovella, Member, IEEE, and Azer Bestavros, Member,

More information

Applying Active Queue Management to Link Layer Buffers for Real-time Traffic over Third Generation Wireless Networks

Applying Active Queue Management to Link Layer Buffers for Real-time Traffic over Third Generation Wireless Networks Applying Active Queue Management to Link Layer Buffers for Real-time Traffic over Third Generation Wireless Networks Jian Chen and Victor C.M. Leung Department of Electrical and Computer Engineering The

More information

NETWORK BURST MONITORING AND DETECTION BASED ON FRACTAL DIMENSION WITH ADAPTIVE TIME-SLOT MONITORING MECHANISM

NETWORK BURST MONITORING AND DETECTION BASED ON FRACTAL DIMENSION WITH ADAPTIVE TIME-SLOT MONITORING MECHANISM 686 Journal of Marine Science and Technology, Vol. 21, No. 6, pp.686-694 (213) DOI: 1.6119/JMST-13-516-1 NETWORK BURST MONITORING AND DETECTION BASED ON FRACTAL DIMENSION WITH ADAPTIVE TIME-SLOT MONITORING

More information

Time Series Analysis of Network Traffic

Time Series Analysis of Network Traffic Time Series Analysis of Network Traffic Cyriac James IIT MADRAS February 9, 211 Cyriac James (IIT MADRAS) February 9, 211 1 / 31 Outline of the presentation Background Motivation for the Work Network Trace

More information

Assignment #2 for Computer Networks

Assignment #2 for Computer Networks Assignment # for Computer Networks Savvas C. Nikiforou Department of Computer Science and Engineering University of South Florida Tampa, FL 6 Abstract The purpose of this assignment is to compare the queueing

More information

Low-rate TCP-targeted Denial of Service Attack Defense

Low-rate TCP-targeted Denial of Service Attack Defense Low-rate TCP-targeted Denial of Service Attack Defense Johnny Tsao Petros Efstathopoulos University of California, Los Angeles, Computer Science Department Los Angeles, CA E-mail: {johnny5t, pefstath}@cs.ucla.edu

More information

Capturing the Complete Multifractal Characteristics of Network Traffic

Capturing the Complete Multifractal Characteristics of Network Traffic Capturing the Complete Multifractal Characteristics of Network Traffic Trang Dinh Dang, Sándor Molnár, István Maricza High Speed Networks Laboratory, Dept. of Telecommunications & Telematics Budapest University

More information

MODELLING AND FORECASTING OF CLOUD DATA WAREHOUSING LOAD

MODELLING AND FORECASTING OF CLOUD DATA WAREHOUSING LOAD STaaS, modelling, cloud computing Rostyslav STRUBYTSKYI * MODELLING AND FORECASTING OF CLOUD DATA WAREHOUSING LOAD Abstract Cloud data storages in their internal structure are not using their full potential

More information

TCP in Wireless Mobile Networks

TCP in Wireless Mobile Networks TCP in Wireless Mobile Networks 1 Outline Introduction to transport layer Introduction to TCP (Internet) congestion control Congestion control in wireless networks 2 Transport Layer v.s. Network Layer

More information

A Compound Model for TCP Connection Arrivals

A Compound Model for TCP Connection Arrivals A Compound Model for TCP Connection Arrivals Carl J Nuzman Dept of Electrical Engineering Princeton University Iraj Saniee Wim Sweldens Alan Weiss Bell Laboratories, Lucent Technologies Murray Hill, NJ

More information

A Passive Method for Estimating End-to-End TCP Packet Loss

A Passive Method for Estimating End-to-End TCP Packet Loss A Passive Method for Estimating End-to-End TCP Packet Loss Peter Benko and Andras Veres Traffic Analysis and Network Performance Laboratory, Ericsson Research, Budapest, Hungary {Peter.Benko, Andras.Veres}@eth.ericsson.se

More information

Flow aware networking for effective quality of service control

Flow aware networking for effective quality of service control IMA Workshop on Scaling 22-24 October 1999 Flow aware networking for effective quality of service control Jim Roberts France Télécom - CNET james.roberts@cnet.francetelecom.fr Centre National d'etudes

More information

Performance Analysis of AQM Schemes in Wired and Wireless Networks based on TCP flow

Performance Analysis of AQM Schemes in Wired and Wireless Networks based on TCP flow International Journal of Soft Computing and Engineering (IJSCE) Performance Analysis of AQM Schemes in Wired and Wireless Networks based on TCP flow Abdullah Al Masud, Hossain Md. Shamim, Amina Akhter

More information

Network Traffic Invariant Characteristics:Metering Aspects

Network Traffic Invariant Characteristics:Metering Aspects etwork Traffic Invariant Characteristics:Metering Aspects Vladimir Zaborovsky, Andrey Rudskoy, Alex Sigalov Politechnical University, Robotics Institute St.Petersburg, Russia; Fractel Inc., USA, E-mail:

More information

Accelerated Simulation Method for Power-law Traffic and Non- FIFO Scheduling

Accelerated Simulation Method for Power-law Traffic and Non- FIFO Scheduling Accelerated Simulation Method for Power-law Traffic and Non- FIF Scheduling Authors: Sharifah H. S. Ariffin and John A. Schormans Department of Electronic Engineering, Queen Mary, University of London,

More information

VoIP Network Dimensioning using Delay and Loss Bounds for Voice and Data Applications

VoIP Network Dimensioning using Delay and Loss Bounds for Voice and Data Applications VoIP Network Dimensioning using Delay and Loss Bounds for Voice and Data Applications Veselin Rakocevic School of Engineering and Mathematical Sciences City University, London, UK V.Rakocevic@city.ac.uk

More information

Comparative Analysis of Congestion Control Algorithms Using ns-2

Comparative Analysis of Congestion Control Algorithms Using ns-2 www.ijcsi.org 89 Comparative Analysis of Congestion Control Algorithms Using ns-2 Sanjeev Patel 1, P. K. Gupta 2, Arjun Garg 3, Prateek Mehrotra 4 and Manish Chhabra 5 1 Deptt. of Computer Sc. & Engg,

More information

Published in: Proceedings of 6th International Computer Engineering Conference (ICENCO)

Published in: Proceedings of 6th International Computer Engineering Conference (ICENCO) Aalborg Universitet Characterization and Modeling of Network Shawky, Ahmed Sherif Mahmoud; Bergheim, Hans ; Ragnarsson, Olafur ; Wranty, Andrzej ; Pedersen, Jens Myrup Published in: Proceedings of 6th

More information

RESEARCH OF THE NETWORK SERVER IN SELF-SIMILAR TRAFFIC ENVIRONMENT

RESEARCH OF THE NETWORK SERVER IN SELF-SIMILAR TRAFFIC ENVIRONMENT RESEARCH OF THE NETWORK SERVER IN SELF-SIMILAR TRAFFIC ENVIRONMENT Sergejs Ilnickis Keywords: network traffic, self-similar traffic, traffic analysis. Abstract - Last scientific publication shows that

More information

Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation

Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation R.Navaneethakrishnan Assistant Professor (SG) Bharathiyar College of Engineering and Technology, Karaikal, India.

More information

Performance Analysis of a VoIP Access Architecture

Performance Analysis of a VoIP Access Architecture Performance Analysis of a VoIP Access Architecture E. Noel AT&T Laboratories Middletown NJ eric.noel@att.com K. W. Tang Dept. of Electrical Engineering SUNY at Stony Brook NY wtang@ece.sunysb.edu Abstract

More information

Master s Thesis. A Study on Active Queue Management Mechanisms for. Internet Routers: Design, Performance Analysis, and.

Master s Thesis. A Study on Active Queue Management Mechanisms for. Internet Routers: Design, Performance Analysis, and. Master s Thesis Title A Study on Active Queue Management Mechanisms for Internet Routers: Design, Performance Analysis, and Parameter Tuning Supervisor Prof. Masayuki Murata Author Tomoya Eguchi February

More information

Analysis of ADSL traffic on an IP backbone link

Analysis of ADSL traffic on an IP backbone link Analysis of ADSL traffic on an IP backbone link Nadia Ben Azzouna and Fabrice Guillemin France Telecom R&D 2, Avenue Pierre Marzin, 223 Lannion, France Abstract Measurements from an Internet backbone link

More information

Network Performance Measurement and Analysis

Network Performance Measurement and Analysis Network Performance Measurement and Analysis Outline Measurement Tools and Techniques Workload generation Analysis Basic statistics Queuing models Simulation CS 640 1 Measurement and Analysis Overview

More information

A Traffic Analysis per Application in a real IP/MPLS Service Provider Network

A Traffic Analysis per Application in a real IP/MPLS Service Provider Network A Traffic Analysis per Application in a real IP/MPLS Service Provider Network Paulo H. P. de Carvalho 1, Priscila Solís Barreto 1,2, Bruno G. Queiroz 1, Breno N. Carneiro 1, Marcio A. de Deus 1 1 Electrical

More information

Using median filtering in active queue management for telecommunication networks

Using median filtering in active queue management for telecommunication networks Using median filtering in active queue management for telecommunication networks Sorin ZOICAN *, Ph.D. Cuvinte cheie. Managementul cozilor de aşteptare, filtru median, probabilitate de rejectare, întârziere.

More information

Seamless Congestion Control over Wired and Wireless IEEE 802.11 Networks

Seamless Congestion Control over Wired and Wireless IEEE 802.11 Networks Seamless Congestion Control over Wired and Wireless IEEE 802.11 Networks Vasilios A. Siris and Despina Triantafyllidou Institute of Computer Science (ICS) Foundation for Research and Technology - Hellas

More information

15-441: Computer Networks Homework 2 Solution

15-441: Computer Networks Homework 2 Solution 5-44: omputer Networks Homework 2 Solution Assigned: September 25, 2002. Due: October 7, 2002 in class. In this homework you will test your understanding of the TP concepts taught in class including flow

More information

Mathematical Modelling of Computer Networks: Part II. Module 1: Network Coding

Mathematical Modelling of Computer Networks: Part II. Module 1: Network Coding Mathematical Modelling of Computer Networks: Part II Module 1: Network Coding Lecture 3: Network coding and TCP 12th November 2013 Laila Daniel and Krishnan Narayanan Dept. of Computer Science, University

More information

An enhanced TCP mechanism Fast-TCP in IP networks with wireless links

An enhanced TCP mechanism Fast-TCP in IP networks with wireless links Wireless Networks 6 (2000) 375 379 375 An enhanced TCP mechanism Fast-TCP in IP networks with wireless links Jian Ma a, Jussi Ruutu b and Jing Wu c a Nokia China R&D Center, No. 10, He Ping Li Dong Jie,

More information

EVALUATION OF LOAD BALANCING ALGORITHMS AND INTERNET TRAFFIC MODELING FOR PERFORMANCE ANALYSIS. Arthur L. Blais

EVALUATION OF LOAD BALANCING ALGORITHMS AND INTERNET TRAFFIC MODELING FOR PERFORMANCE ANALYSIS. Arthur L. Blais EVALUATION OF LOAD BALANCING ALGORITHMS AND INTERNET TRAFFIC MODELING FOR PERFORMANCE ANALYSIS by Arthur L. Blais B.A., California State University, Fullerton, 1982 A thesis submitted to the Graduate Faculty

More information

Simulating Timescale Dynamics of Network Traffic Using Homogeneous Modeling

Simulating Timescale Dynamics of Network Traffic Using Homogeneous Modeling [J. Res. Natl. Inst. Stand. Technol. 111, 227-242 (2006)] Simulating Timescale Dynamics of Network Traffic Using Homogeneous Modeling Volume 111 Number 3 May-June 2006 Jian Yuan Tsinghua University Beijing,

More information

Rate-Based Active Queue Management: A Green Algorithm in Congestion Control

Rate-Based Active Queue Management: A Green Algorithm in Congestion Control Rate-Based Active Queue Management: A Green Algorithm in Congestion Control Balveer Singh #1, Diwakar Saraswat #2 #1 HOD Computer Sc. & Engg. #2 Astt. Prof. Computer Sc. & Engg PKITM Mathura (UP) India

More information

Modeling Heterogeneous Network Traffic in Wavelet Domain

Modeling Heterogeneous Network Traffic in Wavelet Domain 634 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 9, NO. 5, OCTOBER 2001 Modeling Heterogeneous Network Traffic in Wavelet Domain Sheng Ma, Member, IEEE, Chuanyi Ji Abstract Heterogeneous network traffic possesses

More information

A Policy-Based Admission Control Scheme for Voice over IP Networks

A Policy-Based Admission Control Scheme for Voice over IP Networks Journal of Computer Science 5 (11): 817-821, 2009 ISSN 1549-3636 2009 Science Publications A Policy-Based Admission Control Scheme for Voice over IP Networks Sami Alwakeel and Agung Prasetijo Department

More information

Sizing Internet Router Buffers, Active Queue Management, and the Lur e Problem

Sizing Internet Router Buffers, Active Queue Management, and the Lur e Problem Sizing Internet Router Buffers, Active Queue Management, and the Lur e Problem Christopher M. Kellett, Robert N. Shorten, and Douglas J. Leith Abstract Recent work in sizing Internet router buffers has

More information

A Spectrum of TCP-Friendly Window-Based Congestion Control Algorithms

A Spectrum of TCP-Friendly Window-Based Congestion Control Algorithms IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 11, NO. 3, JUNE 2003 341 A Spectrum of TCP-Friendly Window-Based Congestion Control Algorithms Shudong Jin, Liang Guo, Student Member, IEEE, Ibrahim Matta, Member,

More information

Recent Advances in Web System Performance Modeling with Queueing Networks. Author: Nikola Janevski Class: CS 736 Software Performance Engineering

Recent Advances in Web System Performance Modeling with Queueing Networks. Author: Nikola Janevski Class: CS 736 Software Performance Engineering Recent Advances in Web System Performance Modeling with Queueing Networks Author: Nikola Janevski Class: CS 736 Software Performance Engineering 1 How are Web systems different Many users Multi-tier architecture

More information

Robust Router Congestion Control Using Acceptance and Departure Rate Measures

Robust Router Congestion Control Using Acceptance and Departure Rate Measures Robust Router Congestion Control Using Acceptance and Departure Rate Measures Ganesh Gopalakrishnan a, Sneha Kasera b, Catherine Loader c, and Xin Wang b a {ganeshg@microsoft.com}, Microsoft Corporation,

More information

Adaptive DCF of MAC for VoIP services using IEEE 802.11 networks

Adaptive DCF of MAC for VoIP services using IEEE 802.11 networks Adaptive DCF of MAC for VoIP services using IEEE 802.11 networks 1 Mr. Praveen S Patil, 2 Mr. Rabinarayan Panda, 3 Mr. Sunil Kumar R D 1,2,3 Asst. Professor, Department of MCA, The Oxford College of Engineering,

More information

Effects of Filler Traffic In IP Networks. Adam Feldman April 5, 2001 Master s Project

Effects of Filler Traffic In IP Networks. Adam Feldman April 5, 2001 Master s Project Effects of Filler Traffic In IP Networks Adam Feldman April 5, 2001 Master s Project Abstract On the Internet, there is a well-documented requirement that much more bandwidth be available than is used

More information

Ten Fallacies and Pitfalls on End-to-End Available Bandwidth Estimation

Ten Fallacies and Pitfalls on End-to-End Available Bandwidth Estimation Ten Fallacies and Pitfalls on End-to-End Available Bandwidth Estimation Manish Jain Georgia Tech jain@cc.gatech.edu Constantinos Dovrolis Georgia Tech dovrolis@cc.gatech.edu ABSTRACT The area of available

More information

The Interaction of Forward Error Correction and Active Queue Management

The Interaction of Forward Error Correction and Active Queue Management The Interaction of Forward Error Correction and Active Queue Management Tigist Alemu, Yvan Calas, and Alain Jean-Marie LIRMM UMR 5506 CNRS and University of Montpellier II 161, Rue Ada, 34392 Montpellier

More information

A Flow- and Packet-level Model of the Internet

A Flow- and Packet-level Model of the Internet A Flow- and Packet-level Model of the Internet Ramji Venkataramanan Dept. of Electrical Engineering Yale University, USA Email: ramji.venkataramanan@yale.edu Min-Wook Jeong, Balaji Prabhakar Dept. of Electrical

More information

Adaptive Coding and Packet Rates for TCP-Friendly VoIP Flows

Adaptive Coding and Packet Rates for TCP-Friendly VoIP Flows Adaptive Coding and Packet Rates for TCP-Friendly VoIP Flows C. Mahlo, C. Hoene, A. Rostami, A. Wolisz Technical University of Berlin, TKN, Sekr. FT 5-2 Einsteinufer 25, 10587 Berlin, Germany. Emails:

More information

CASCADE models or multiplicative processes make especially

CASCADE models or multiplicative processes make especially IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 45, NO. 3, APRIL 1999 971 Scaling Analysis of Conservative Cascades, with Applications to Network Traffic A. C. Gilbert, W. Willinger, Member, IEEE, and A.

More information

Oscillations of the Sending Window in Compound TCP

Oscillations of the Sending Window in Compound TCP Oscillations of the Sending Window in Compound TCP Alberto Blanc 1, Denis Collange 1, and Konstantin Avrachenkov 2 1 Orange Labs, 905 rue Albert Einstein, 06921 Sophia Antipolis, France 2 I.N.R.I.A. 2004

More information

Network congestion, its control and avoidance

Network congestion, its control and avoidance MUHAMMAD SALEH SHAH*, ASIM IMDAD WAGAN**, AND MUKHTIAR ALI UNAR*** RECEIVED ON 05.10.2013 ACCEPTED ON 09.01.2014 ABSTRACT Recent years have seen an increasing interest in the design of AQM (Active Queue

More information

Lecture 8 Performance Measurements and Metrics. Performance Metrics. Outline. Performance Metrics. Performance Metrics Performance Measurements

Lecture 8 Performance Measurements and Metrics. Performance Metrics. Outline. Performance Metrics. Performance Metrics Performance Measurements Outline Lecture 8 Performance Measurements and Metrics Performance Metrics Performance Measurements Kurose-Ross: 1.2-1.4 (Hassan-Jain: Chapter 3 Performance Measurement of TCP/IP Networks ) 2010-02-17

More information

Analysis of VDI Workload Characteristics

Analysis of VDI Workload Characteristics Analysis of VDI Workload Characteristics Jeongsook Park, Youngchul Kim and Youngkyun Kim Electronics and Telecommunications Research Institute 218 Gajeong-ro, Yuseong-gu, Daejeon, 305-700, Republic of

More information

SIMULATION AND ANALYSIS OF QUALITY OF SERVICE PARAMETERS IN IP NETWORKS WITH VIDEO TRAFFIC

SIMULATION AND ANALYSIS OF QUALITY OF SERVICE PARAMETERS IN IP NETWORKS WITH VIDEO TRAFFIC SIMULATION AND ANALYSIS OF QUALITY OF SERVICE PARAMETERS IN IP NETWORKS WITH VIDEO TRAFFIC by Bruce Chen THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF BACHELOR OF APPLIED

More information

NETWORK REQUIREMENTS FOR HIGH-SPEED REAL-TIME MULTIMEDIA DATA STREAMS

NETWORK REQUIREMENTS FOR HIGH-SPEED REAL-TIME MULTIMEDIA DATA STREAMS NETWORK REQUIREMENTS FOR HIGH-SPEED REAL-TIME MULTIMEDIA DATA STREAMS Andrei Sukhov 1), Prasad Calyam 2), Warren Daly 3), Alexander Iliin 4) 1) Laboratory of Network Technologies, Samara Academy of Transport

More information

On the Characteristics and Origins of Internet Flow Rates

On the Characteristics and Origins of Internet Flow Rates On the Characteristics and Origins of Internet Flow Rates Yin Zhang Lee Breslau AT&T Labs Research {yzhang,breslau}@research.att.com Vern Paxson Scott Shenker International Computer Science Institute {vern,shenker}@icsi.berkeley.edu

More information

TCP/IP Performance with Random Loss and Bidirectional Congestion

TCP/IP Performance with Random Loss and Bidirectional Congestion IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 8, NO. 5, OCTOBER 2000 541 TCP/IP Performance with Random Loss and Bidirectional Congestion T. V. Lakshman, Senior Member, IEEE, Upamanyu Madhow, Senior Member,

More information

Transport layer issues in ad hoc wireless networks Dmitrij Lagutin, dlagutin@cc.hut.fi

Transport layer issues in ad hoc wireless networks Dmitrij Lagutin, dlagutin@cc.hut.fi Transport layer issues in ad hoc wireless networks Dmitrij Lagutin, dlagutin@cc.hut.fi 1. Introduction Ad hoc wireless networks pose a big challenge for transport layer protocol and transport layer protocols

More information

Analyzing Marking Mod RED Active Queue Management Scheme on TCP Applications

Analyzing Marking Mod RED Active Queue Management Scheme on TCP Applications 212 International Conference on Information and Network Technology (ICINT 212) IPCSIT vol. 7 (212) (212) IACSIT Press, Singapore Analyzing Marking Active Queue Management Scheme on TCP Applications G.A.

More information

AN IMPROVED SNOOP FOR TCP RENO AND TCP SACK IN WIRED-CUM- WIRELESS NETWORKS

AN IMPROVED SNOOP FOR TCP RENO AND TCP SACK IN WIRED-CUM- WIRELESS NETWORKS AN IMPROVED SNOOP FOR TCP RENO AND TCP SACK IN WIRED-CUM- WIRELESS NETWORKS Srikanth Tiyyagura Department of Computer Science and Engineering JNTUA College of Engg., pulivendula, Andhra Pradesh, India.

More information

Bandwidth Allocation under End-to-End Percentile Delay Bounds

Bandwidth Allocation under End-to-End Percentile Delay Bounds Bandwidth Allocation under End-to-End Percentile Delay Bounds Bushra Anjum, Harry Perros, Xenia Mountrouidou, Kimon Kontovasilis North Carolina State University, Computer Science Department, Raleigh, NC

More information

Multiple TFRC Connections Based Rate Control for Wireless Networks

Multiple TFRC Connections Based Rate Control for Wireless Networks Multiple TFRC Connections Based Rate Control for Wireless Networks Minghua Chen, Student Member, IEEE, and Avideh Zakhor Fellow, IEEE Abstract Rate control is an important issue in video streaming applications

More information

International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July-2015 1169 ISSN 2229-5518

International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July-2015 1169 ISSN 2229-5518 International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July-2015 1169 Comparison of TCP I-Vegas with TCP Vegas in Wired-cum-Wireless Network Nitin Jain & Dr. Neelam Srivastava Abstract

More information

Modeling and Analysis of Wireless LAN Traffic *

Modeling and Analysis of Wireless LAN Traffic * JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 5, 1783-1801 (009) Modeling and Analysis of Wireless LAN Traffic * DASHDORJ YAMKHIN AND YOUJIP WON + Department of Electronics and Computer Engineering Hanyang

More information

Bandwidth Measurement in Wireless Networks

Bandwidth Measurement in Wireless Networks Bandwidth Measurement in Wireless Networks Andreas Johnsson, Bob Melander, and Mats Björkman {andreas.johnsson, bob.melander, mats.bjorkman}@mdh.se The Department of Computer Science and Engineering Mälardalen

More information

Active Queue Management

Active Queue Management Course of Multimedia Internet (Sub-course Reti Internet Multimediali ), AA 2010-2011 Prof. 6. Active queue management Pag. 1 Active Queue Management Active Queue Management (AQM) is a feature that can

More information

Multi-service Load Balancing in a Heterogeneous Network with Vertical Handover

Multi-service Load Balancing in a Heterogeneous Network with Vertical Handover 1 Multi-service Load Balancing in a Heterogeneous Network with Vertical Handover Jie Xu, Member, IEEE, Yuming Jiang, Member, IEEE, and Andrew Perkis, Member, IEEE Abstract In this paper we investigate

More information

Corrected Diffusion Approximations for the Maximum of Heavy-Tailed Random Walk

Corrected Diffusion Approximations for the Maximum of Heavy-Tailed Random Walk Corrected Diffusion Approximations for the Maximum of Heavy-Tailed Random Walk Jose Blanchet and Peter Glynn December, 2003. Let (X n : n 1) be a sequence of independent and identically distributed random

More information

Linking Network Usage Patterns to Traffic Gaussianity Fit

Linking Network Usage Patterns to Traffic Gaussianity Fit Linking Network Usage Patterns to Traffic Gaussianity Fit Ricardo de O. Schmidt, Ramin Sadre, Nikolay Melnikov, Jürgen Schönwälder, Aiko Pras University of Twente, The Netherlands Email: {r.schmidt,a.pras}@utwente.nl

More information

On the Performance of Bandwidth Allocation Strategies for Interconnecting ATM and Connectionless Networks

On the Performance of Bandwidth Allocation Strategies for Interconnecting ATM and Connectionless Networks On the Performance of Bandwidth Allocation Strategies for Interconnecting ATM and Connectionless Networks Edward Chan, Victor C. S. Lee Department of Computer Science, City University of Hong Kong e-mail:

More information

Application Layer Traffic Analysis of a Peer-to-Peer System

Application Layer Traffic Analysis of a Peer-to-Peer System Application Layer Traffic Analysis of a Peer-to-Peer System Dietmar Tutsch Technical University Berlin Institute of Computer Engineering and Microelectronics Berlin, Germany DietmarT@cs.tu-berlin.de Gilbert

More information

The Dependence of Internet User Traffic Characteristics on Access Speed

The Dependence of Internet User Traffic Characteristics on Access Speed University of Würzburg Institute of Computer Science Research Report Series The Dependence of Internet User Traffic Characteristics on Access Speed Norbert Vicari*, Stefan Köhler* and Joachim Charzinski

More information

Adaptive Bandwidth Allocation Method for Long Range Dependence Traffic

Adaptive Bandwidth Allocation Method for Long Range Dependence Traffic Adaptive Bandwidth Allocation Method for Long Range Dependence Traffic Bong Joo Kim and Gang Uk Hwang Division of Applied Mathematics and Telecommunication Program Korea Advanced Institute of Science and

More information

How to generate realistic network traffic?

How to generate realistic network traffic? How to generate realistic network traffic? Antoine Varet, Nicolas Larrieu To cite this version: Antoine Varet, Nicolas Larrieu. How to generate realistic network traffic?. IEEE COMPSAC 2014, 38th Annual

More information

Conversion range λ 1 λ n*n

Conversion range λ 1 λ n*n WDM Fiber Delay Line Buer Control for Optical Packet Switching An Ge a, Ljubisa Tancevski a, Gerardo Castanon a and Lakshman S. Tamil b a Corporate Research Center, Alcatel USA, Richardson, TX 7508 b Yotta

More information

TCP and UDP Performance for Internet over Optical Packet-Switched Networks

TCP and UDP Performance for Internet over Optical Packet-Switched Networks TCP and UDP Performance for Internet over Optical Packet-Switched Networks Jingyi He S-H Gary Chan Department of Electrical and Electronic Engineering Department of Computer Science Hong Kong University

More information

IEEE --- 2005 International Conference on Emerging Technologies September 17-18, Islamabad

IEEE --- 2005 International Conference on Emerging Technologies September 17-18, Islamabad Sajjad Ali Mushtaq 1 & Azhar A. Rizvi 2 CIIT Abbottabad Pakistan 1 & Department of Electronics, Quaid-I-Azam University Islamabad, Pakistan 2. E-mail: alisajjad_mushtaq@yahoo.com 1 & azhar@qau.edu.pk 2

More information

Traffic analysis and network bandwidth provisioning tools for academic information networks

Traffic analysis and network bandwidth provisioning tools for academic information networks Progress in Informatics, No. 1, pp. 83-91, (005) 83 R&D Project Report Traffic analysis and network bandwidth provisioning tools for academic information networks Shunji Abe 1, Toru Hasegawa, Shoichiro

More information

Outline. TCP connection setup/data transfer. 15-441 Computer Networking. TCP Reliability. Congestion sources and collapse. Congestion control basics

Outline. TCP connection setup/data transfer. 15-441 Computer Networking. TCP Reliability. Congestion sources and collapse. Congestion control basics Outline 15-441 Computer Networking Lecture 8 TCP & Congestion Control TCP connection setup/data transfer TCP Reliability Congestion sources and collapse Congestion control basics Lecture 8: 09-23-2002

More information

Performance Analysis and Software Optimization on Systems Using the LAN91C111

Performance Analysis and Software Optimization on Systems Using the LAN91C111 AN 10.12 Performance Analysis and Software Optimization on Systems Using the LAN91C111 1 Introduction This application note describes one approach to analyzing the performance of a LAN91C111 implementation

More information

An Overview and Comparison of Analytical TCP Models

An Overview and Comparison of Analytical TCP Models An Overview and Comparison of Analytical TCP Models Inas Khalifa and Ljiljana Trajkovic Communication Networks Laboratory http://www.ensc.sfu.ca/research/cnl School of Engineering Science Simon Fraser

More information

On Packet Marking Function of Active Queue Management Mechanism: Should It Be Linear, Concave, or Convex?

On Packet Marking Function of Active Queue Management Mechanism: Should It Be Linear, Concave, or Convex? On Packet Marking Function of Active Queue Management Mechanism: Should It Be Linear, Concave, or Convex? Hiroyuki Ohsaki and Masayuki Murata Graduate School of Information Science and Technology Osaka

More information

A Survey on Congestion Control Mechanisms for Performance Improvement of TCP

A Survey on Congestion Control Mechanisms for Performance Improvement of TCP A Survey on Congestion Control Mechanisms for Performance Improvement of TCP Shital N. Karande Department of Computer Science Engineering, VIT, Pune, Maharashtra, India Sanjesh S. Pawale Department of

More information

CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING

CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING CHAPTER 6 CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING 6.1 INTRODUCTION The technical challenges in WMNs are load balancing, optimal routing, fairness, network auto-configuration and mobility

More information

Master s Thesis. Design, Implementation and Evaluation of

Master s Thesis. Design, Implementation and Evaluation of Master s Thesis Title Design, Implementation and Evaluation of Scalable Resource Management System for Internet Servers Supervisor Prof. Masayuki Murata Author Takuya Okamoto February, 2003 Department

More information

Performance improvement of active queue management with per-flow scheduling

Performance improvement of active queue management with per-flow scheduling Performance improvement of active queue management with per-flow scheduling Masayoshi Nabeshima, Kouji Yata NTT Cyber Solutions Laboratories, NTT Corporation 1-1 Hikari-no-oka Yokosuka-shi Kanagawa 239

More information

A Short Look on Power Saving Mechanisms in the Wireless LAN Standard Draft IEEE 802.11

A Short Look on Power Saving Mechanisms in the Wireless LAN Standard Draft IEEE 802.11 A Short Look on Power Saving Mechanisms in the Wireless LAN Standard Draft IEEE 802.11 Christian Röhl, Hagen Woesner, Adam Wolisz * Technical University Berlin Telecommunication Networks Group {roehl,

More information

MODELLING INTERNET PACKET TRAFFIC CONGESTION

MODELLING INTERNET PACKET TRAFFIC CONGESTION MODELLING INTERNET PACKET TRAFFIC CONGESTION DAVID ARROWSMITH, R. Mondragon, J. Pitts, M. Woolf MATHEMATICAL SCIENCES & ELECTRONIC ENGINEERING QUEEN MARY, UNIVERSITY OF LONDON LONDON E 4NS, UK www.maths.qmul.ac.uk/~arrow/ieeebangkok.pdf

More information

Generation of High Bandwidth Network Traffic Traces

Generation of High Bandwidth Network Traffic Traces Generation of High Bandwidth Network Traffic Traces Purushotham Kamath, Kun-chan Lan, John Heidemann, Joe Bannister and Joe Touch University of Southern California Information Sciences Institute Los Angeles,

More information

Optimal Bandwidth Monitoring. Y.Yu, I.Cheng and A.Basu Department of Computing Science U. of Alberta

Optimal Bandwidth Monitoring. Y.Yu, I.Cheng and A.Basu Department of Computing Science U. of Alberta Optimal Bandwidth Monitoring Y.Yu, I.Cheng and A.Basu Department of Computing Science U. of Alberta Outline Introduction The problem and objectives The Bandwidth Estimation Algorithm Simulation Results

More information