Large Scale Networks by Dobrescu, Radu -- Read -- Imperial Library of Trantor

Index

Cover Half Title Title Page Copyright Page Table of Contents Preface Authors Chapter 1 State of the art and trends in information networks modeling

1.1 Self-similarity and fractals in traffic modeling

1.1.1 Building of a framework 1.1.2 Mathematical background of self-similar processes

1.1.2.1 Stable distributions and semistable processes 1.1.2.2 First empirical processes with power-law statistics 1.1.2.3 From semistability to self-similarity 1.1.2.4 Self-similarity versus long-range dependence 1.1.2.5 Self-similarity discovered in network traffic

1.2 Models of complex networks 1.3 Scale-free networks

1.3.1 Basic properties of SFNs 1.3.2 Distances and bounds in SFN

1.4 Current trends in traffic flow and complex networks modeling

Chapter 2 Flow traffic models

2.1 Background in traffic modeling

2.1.1 Definition of the informational traffic 2.1.2 Internet teletraffic modeling

2.1.2.1 Introduction in teletraffic theory 2.1.2.2 Basic concepts of teletraffic theory 2.1.2.3 Teletraffic techniques

2.1.3 Internet teletraffic engineering

2.1.3.1 Big bandwidth philosophy 2.1.3.2 Managed bandwidth philosophy 2.1.3.3 Open-loop control of stream traffic 2.1.3.4 Closed-loop control of elastic traffic

2.1.4 Internet traffic times series modeling

2.2 Renewal traffic models

2.2.1 Poisson processes 2.2.2 Bernoulli processes 2.2.3 Phase-type renewal processes

2.3 Markov traffic models

2.3.1 Markov-modulated traffic models 2.3.2 Markov-modulated Poisson process 2.3.3 Transition-modulated processes

2.4 Fluid traffic models 2.5 Autoregressive traffic models

2.5.1 Linear autoregressive models (AR) 2.5.2 Moving average series (MA) models 2.5.3 Autoregressive moving average series (ARMA) models 2.5.4 Integrated ARIMA models 2.5.5 FARIMA models

2.6 TES traffic models

2.6.1 TES processes 2.6.2 Empirical TES methodology

2.7 Self-similar traffic models

Chapter 3 Self-similarity in traffic

3.1 Self-similar traffic and network performance

3.1.1 Quality of service and resource allocation 3.1.2 Concept of self-similarity 3.1.3 Effects of self-similarity on network performance

3.2 Mathematics of self-similar processes

3.2.1 Stationary random processes 3.2.2 Continuous time self-similar processes 3.2.3 Discrete time self-similar processes 3.2.4 Properties of the fractal processes

3.2.4.1 Long-range dependence 3.2.4.2 Slowly decaying dispersion 3.2.4.3 Heavy-tailed distributions 3.2.4.4 1/f noise 3.2.4.5 Fractal dimension 3.2.4.6 SRD versus LRD: Conclusion

3.3 Self-similar traffic modeling

3.3.1 Single-source traffic models

3.3.1.1 Pareto distribution 3.3.1.2 Log-normal distribution

3.3.2 Aggregate traffic models

3.3.2.1 ON–OFF models 3.3.2.2 Fractional Brownian motion 3.3.2.3 Fractional Gaussian noise 3.3.2.4 Fractional ARIMA processes (FARIMA) 3.3.2.5 Chaotic deterministic maps

3.3.3 Procedures for synthetic self-similar traffic generation

3.3.3.1 ON–OFF method 3.3.3.2 Norros method 3.3.3.3 M/G/∞ queue method 3.3.3.4 Random midpoint displacement (RMD) method 3.3.3.5 Wavelet transform-based method 3.3.3.6 Successive random addition (SRA)

3.3.4 Fast Fourier transform (FFT) method

3.4 Evidence of self-similarity in real traffic

3.4.1 Rescaled range method 3.4.2 Dispersion–time analysis 3.4.3 Periodogram method 3.4.4 Whittle estimator 3.4.5 Wavelet-based method

3.5 Application specific models

3.5.1 Internet application-specific traffic models

3.5.1.1 Network traffic 3.5.1.2 Web traffic 3.5.1.3 Peer-to-peer traffic 3.5.1.4 Video

3.5.2 Models for TCP flows

3.5.2.1 TCP flows with congestion avoidance 3.5.2.2 TCP flows with active queue management

Chapter 4 Topological models of complex networks

4.1 Topology of large-scale networks

4.1.1 World Wide Web 4.1.2 The Internet

4.2 Main approaches on networks topology modeling

4.2.1 Random graph theory 4.2.2 Small-world networks

4.2.2.1 Average trajectory length 4.2.2.2 Clustering coefficient 4.2.2.3 Degree of distribution 4.2.2.4 Spectral properties

4.2.3 The scale-free (SF) model

4.2.3.1 Definition of the scale-free network (SFN) model 4.2.3.2 General SFN properties 4.2.3.3 Diameter of scale-free networks 4.2.3.4 Average trajectory length 4.2.3.5 Node degree correlation 4.2.3.6 Clustering coefficient

4.3 Internet traffic simulation with scale-free network models

4.3.1 Choosing a model 4.3.2 Proposed Internet model

4.3.2.1 Scale-free network design algorithm 4.3.2.2 Traffic generation 4.3.2.3 Single-CPU simulation 4.3.2.4 Multi-CPU simulations 4.3.2.5 Cluster description 4.3.2.6 Network splitting

4.3.3 Simulation results and discussion

4.4 Improvement of Internet traffic QoS performance

4.4.1 Quality of service (QoS) requirements in highspeed networks 4.4.2 Technologies for high-speed networks

4.4.2.1 Characteristics of OBS networks 4.4.2.2 Burst assembly schemes

4.4.3 Traffic-smoothing using burst assembly schemes

4.4.3.1 Analysis of assembled traffic 4.4.3.2 Description of the proposed burst assembly algorithm

4.4.4 Simulation results

4.5 Fractal approaches in Internet traffic simulation

4.5.1 Using parallel discrete-event simulations (PDES) for scalability evidence in large-scale networks modeling

4.5.1.1 Parallel platform model 4.5.1.2 Share-scheduling policy for parallel processing

4.5.2 Evidence of attractors in large-scale networks models

Chapter 5 Topology and traffic simulations in complex networks

5.1 Example of building and simulating a network

5.1.1 Simple simulation example

5.2 Construction of complex network topologies

5.2.1 Construction of a random network 5.2.2 Construction of a small-world network 5.2.3 Construction of a scale-free network

5.3 Analyses and topological comparisons of complex networks 5.4 Self-similar traffic simulation 5.5 Traffic simulation on combined topologies of networks and traffic sources

5.5.1 Details on the used topologies and traffic sources 5.5.2 Hurst parameter estimation results 5.5.3 Influence of topology upon the traffic

Chapter 6 Case studies

6.1 Hurst exponent analysis on real traffic

6.1.1 Traffic capture 6.1.2 Graphical estimators representation

6.2 Inferring statistical characteristics as an indirect measure of the quality of service

6.2.1 Defining an inference model 6.2.2 Highlighting network similarity 6.2.3 Case study: Interdomain characteristic interference

6.3 Modeling nonlinear phenomena in complex networks and detecting traffic anomalies

6.3.1 Introduction 6.3.2 Self-similarity characteristic of the informational traffic in networks

6.3.2.1 SS: Self-similar processes

6.3.3 Using similarity in network management

6.3.3.1 Related work on anomalies detection methods 6.3.3.2 Anomaly detection using statistical analysis of SNMP–MIB 6.3.3.3 Subnetwork similarity testing

6.3.4 Test platform and processing procedure for traffic analysis

6.3.4.1 Traffic generation 6.3.4.2 Cluster description 6.3.4.3 Network federalization

6.3.5 Discussion on experimental results of case studies

6.3.5.1 Implementing and testing an SFN model for Internet 6.3.5.2 Detecting network anomalies 6.3.5.3 Overall conclusions

6.3.6 Recent trends in traffic self-similarity assessment for Cloud data modeling

6.3.6.1 Anomalies detection in Cloud 6.3.6.2 Combining filtering and statistical methods for anomaly detection

6.4 Optimization of quality of services by monitoring Cloud traffic

6.4.1 Monitoring the dynamics of network traffic in Cloud

6.4.1.1 Prediction algorithm 6.4.1.2 Algorithm for capturing cross-deployment workload changes 6.4.1.3 Provisioning planning algorithm

6.4.2 Coping with traffic uncertainty for load balancing in Cloud 6.4.3 Wide-area data analysis for detection changes in traffic patterns

6.4.3.1 WDAS design 6.4.3.2 Aggregation 6.4.3.3 Degradation

6.4.4 Monitoring Cloud services using NetFlow standard

6.4.4.1 Nonsampled flow data 6.4.4.2 Packet-sampled flow data

6.4.5 Implementing Cloud services in the automation domain

6.5 Developing and validating strategies for traffic monitoring on RLS models

6.5.1 Simulation framework

6.5.1.1 Network topology 6.5.1.2 Traffic sources 6.5.1.3 Simulation code

6.5.2 Algorithms ran in simulation

6.5.2.1 DropTail 6.5.2.2 DRR 6.5.2.3 FQ 6.5.2.4 SFQ 6.5.2.5 RED

6.5.3 Performance analysis

6.5.3.1 Transfer rate analysis 6.5.3.2 UDP delay analysis 6.5.3.3 TCP delay analysis

6.6 Conclusions

Bibliography Index

← Prev
Back
Next →

← Prev
Back
Next →