内容説明
is the first book to set out a range of machine learning methods for efficient resource management in a large distributed network of clouds.
predictive analytics is an integral part of efficient cloud resource management, and this book gives a future research direction to researchers in this domain.
it is written by leading international researchers.
目次
List of Figures
List of Tables
Preface
Author Bios
Abbreviations
Introduction
1.1 CLOUD COMPUTING
1.2 CLOUD MANAGEMENT
1.2.1 Workload Forecasting
1.2.2 Load Balancing
1.3 MACHINE LEARNING
1.3.1 Artificial Neural Network
1.3.2 Metaheuristic Optimization Algorithms
1.3.3 Time Series Analysis
1.4 WORKLOAD TRACES
1.5 EXPERIMENTAL SETUP & EVALUATION METRICS
1.6 STATISTICAL TESTS
1.6.1 Wilcoxon Signed-Rank Test
1.6.2 Friedman Test
1.6.3 Finner Test
Time Series Models
2.1 AUTOREGRESSION
2.2 MOVING AVERAGE
2.3 AUTOREGRESSIVE MOVING AVERAGE
2.4 AUTOREGRESSIVE INTEGRATED MOVING AVERAGE
2.5 EXPONENTIAL SMOOTHING
2.6 EXPERIMENTAL ANALYSIS
2.6.1 Forecast Evaluation
2.6.2 Statistical Analysis
Error Preventive Time Series Models
3.1 ERROR PREVENTION SCHEME
3.2 PREDICTIONS IN ERROR RANGE
3.3 MAGNITUDE OF PREDICTIONS
3.4 ERROR PREVENTIVE TIME SERIES MODELS
3.4.1 Error Preventive Autoregressive Moving Average
3.4.2 Error Preventive Auto Regressive Integrated Moving Average
3.4.3 Error Preventive Exponential Smoothing
3.5 PERFORMANCE EVALUATION
3.5.1 Comparative Analysis
3.5.2 Statistical Analysis
Metaheuristic Optimization Algorithms
4.1 SWARM INTELLIGENCE ALGORITHMS IN PREDICTIVE MODEL
4.1.1 Particle Swarm Optimization
4.1.2 Firefly Search Algorithm
4.2 EVOLUTIONARY ALGORITHMS IN PREDICTIVE MODEL
4.2.1 Genetic Algorithm
4.2.2 Differential Evolution
4.3 NATURE INSPIRED ALGORITHMS IN PREDICTIVE MODEL
4.3.1 Harmony Search
4.3.2 Teaching Learning Based Optimization
4.4 PHYSICS INSPIRED ALGORITHMS IN PREDICTIVE MODEL
4.4.1 Gravitational Search Algorithm
4.4.2 Blackhole Algorithm
4.5 STATISTICAL PERFORMANCE ASSESSMENT
Evolutionary Neural Networks
5.1 NEURAL NETWORK PREDICTION FRAMEWORK DESIGN
5.2 NETWORK LEARNING
5.3 RECOMBINATION OPERATOR STRATEGY LEARNING
5.3.1 Mutation Operator
5.3.1.1 DE/current to best/1
5.3.1.2 DE/best/1
5.3.1.3 DE/rand/1
5.3.2 Crossover Operator
5.3.2.1 Ring Crossover
5.3.2.2 Heuristic Crossover
5.3.2.3 Uniform Crossover
5.3.3 Operator Learning Process
5.4 ALGORITHMS AND ANALYSIS
5.5 FORECAST ASSESSMENT
5.5.1 Short Term Forecast
5.5.2 Long Term Forecast
5.6 COMPARATIVE ANALYSIS
Self Directed Learning
6.1 NON-DIRECTED LEARNING BASED FRAMEWORK
6.1.1 Non-Directed Learning
6.2 SELF-DIRECTED LEARNING BASED FRAMEWORK
6.2.1 Self Directed Learning
6.2.2 Cluster Based Learning
6.2.3 Complexity analysis
6.3 FORECAST ASSESSMENT
6.3.1 Short Term Forecast
6.3.1.1 Web Server Workloads
6.3.1.2 Cloud Workloads
6.4 LONG TERM FORECAST
6.4.0.1 Web Server Workloads
6.4.0.2 Cloud Workloads
6.5 COMPARATIVE & STATISTICAL ANALYSIS
Ensemble Learning
7.1 EXTREME LEARNING MACHINE
7.2 WORKLOAD DECOMPOSITION PREDICTIVE FRAMEWORK
7.2.1 Framework Design
7.3 ELM ENSEMBLE PREDICTIVE FRAMEWORK
7.3.1 Ensemble Learning
7.3.2 Expert Architecture Learning
7.3.3 Expert Weight Allocation
7.4 SHORT TERM FORECAST EVALUATION
7.5 LONG TERM FORECAST EVALUATION
7.6 COMPARATIVE ANALYSIS
Load Balancing
8.1 MULTI-OBJECTIVE OPTIMIZATION
8.2 RESOURCE EFFICIENT LOAD BALANCING FRAMEWORK
8.3 SECURE AND ENERGY AWARE LOAD BALANCING FRAMEWORK
8.3.1 Side Channel Attacks
8.3.2 Ternary Objective VM Placement
8.4 SIMULATION SETUP
8.5 HOMOGENEOUS VM PLACEMENT ANALYSIS
8.6 HETEROGENEOUS VM PLACEMENT ANALYSIS
Bibliography
Index
「Nielsen BookData」 より