Machine Learning Techniques and Analytics For Cloud Security Advances in Learning Analytics For Intelligent Cloud Iot Systems 1St Edition Chakraborty

Full download test bank at ebookmeta.
com
Machine Learning Techniques and Analytics for

Cloud Security Advances in Learning Analytics for
Intelligent Cloud IoT Systems 1st Edition
Chakraborty
For dowload this book click LINK or Button below
https://ebookmeta.com/product/machine-learning-
techniques-and-analytics-for-cloud-security-
advances-in-learning-analytics-for-intelligent-
cloud-iot-systems-1st-edition-chakraborty/
OR CLICK BUTTON
DOWLOAD EBOOK
Download More ebooks from https://ebookmeta.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Machine Learning Approach for Cloud Data Analytics in

IoT 1st Edition Sachi Nandan Mohanty (Editor)
https://ebookmeta.com/product/machine-learning-approach-for-
cloud-data-analytics-in-iot-1st-edition-sachi-nandan-mohanty-
editor/
Machine Learning for Intelligent Multimedia Analytics

Techniques and Applications Pardeep Kumar Amit Kumar
Singh Eds
https://ebookmeta.com/product/machine-learning-for-intelligent-
multimedia-analytics-techniques-and-applications-pardeep-kumar-
amit-kumar-singh-eds/
Machine Learning and IoT for Intelligent Systems and

Smart Applications 1st Edition
https://ebookmeta.com/product/machine-learning-and-iot-for-
intelligent-systems-and-smart-applications-1st-edition/
Deep Learning Approaches to Cloud Security Deep

Learning Approaches for Cloud Security 1st Edition
https://ebookmeta.com/product/deep-learning-approaches-to-cloud-
security-deep-learning-approaches-for-cloud-security-1st-edition/
Data-Driven Mining, Learning and Analytics for Secured
Smart Cities: Trends and Advances (Advanced Sciences
and Technologies for Security Applications) Chinmay
Chakraborty
https://ebookmeta.com/product/data-driven-mining-learning-and-
analytics-for-secured-smart-cities-trends-and-advances-advanced-
sciences-and-technologies-for-security-applications-chinmay-
chakraborty/
Machine Learning for Cloud Management 1st Edition Kumar
https://ebookmeta.com/product/machine-learning-for-cloud-
management-1st-edition-kumar/
Machine Learning for Intelligent Multimedia Analytics

Techniques and Applications Studies in Big Data 82
Pardeep Kumar Editor Amit Kumar Singh Editor
https://ebookmeta.com/product/machine-learning-for-intelligent-
multimedia-analytics-techniques-and-applications-studies-in-big-
data-82-pardeep-kumar-editor-amit-kumar-singh-editor/
Advances in Machine Learning Deep Learning based

Technologies Selected Papers in Honour of Professor
Nikolaos G Bourbakis Vol 2 Learning and Analytics in
Intelligent Systems 23 George A. Tsihrintzis (Editor)
https://ebookmeta.com/product/advances-in-machine-learning-deep-
learning-based-technologies-selected-papers-in-honour-of-
professor-nikolaos-g-bourbakis-vol-2-learning-and-analytics-in-
intelligent-systems-23-george-a-tsihrintzis/
Architecting Data and Machine Learning Platforms:

Enable Analytics and AI-Driven Innovation in the Cloud
1st Edition Marco Tranquillin
https://ebookmeta.com/product/architecting-data-and-machine-
learning-platforms-enable-analytics-and-ai-driven-innovation-in-
the-cloud-1st-edition-marco-tranquillin/
Machine Learning Techniques
and Analytics for Cloud Security
Scrivener Publishing
100 Cummings Center, Suite 541J
Beverly, MA 01915-6106
Advances in Learning Analytics for Intelligent Cloud-IoT Systems
Series Editor: Dr. Souvik Pal and Dr. Dac-Nhuong Le
The role of adaptation, learning analytics, computational Intelligence, and data analytics in the field
of cloud-IoT systems is becoming increasingly essential and intertwined. The capability of an
intelligent system depends on various self-decision-making algorithms in IoT devices. IoT-based
smart systems generate a large amount of data (big data) that cannot be processed by traditional data
processing algorithms and applications. Hence, this book series involves different computational
methods incorporated within the system with the help of analytics reasoning and sense-making in big
data, which is centered in the cloud and IoT-enabled environments. The series publishes volumes that
are empirical studies, theoretical and numerical analysis, and novel research findings.
Submission to the series:

Please send proposals to Dr. Souvik Pal, Department of Computer Science and Engineering,
Global Institute of Management and Technology, Krishna Nagar, West Bengal, India.
E-mail: souvikpal22@gmail.com
Publishers at Scrivener
Martin Scrivener (martin@scrivenerpublishing.com)
Phillip Carmical (pcarmical@scrivenerpublishing.com)
Machine Learning Techniques
and Analytics for Cloud Security
Edited by
Rajdeep Chakraborty
Anupam Ghosh
and
Jyotsna Kumar Mandal
This edition first published 2022 by John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA and Scrivener
Publishing LLC, 100 Cummings Center, Suite 541J, Beverly, MA 01915, USA
© 2022 Scrivener Publishing LLC
For more information about Scrivener publications please visit www.scrivenerpublishing.com.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or
by any means, electronic, mechanical, photocopying, recording, or otherwise, except as permitted by law. Advice on how to
obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
Wiley Global Headquarters

111 River Street, Hoboken, NJ 07030, USA
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.
wiley.com.
Limit of Liability/Disclaimer of Warranty

While the publisher and authors have used their best efforts in preparing this work, they make no representations or warran-
ties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including
without limitation any implied warranties of merchant-ability or fitness for a particular purpose. No warranty may be created
or extended by sales representatives, written sales materials, or promotional statements for this work. The fact that an orga-
nization, website, or product is referred to in this work as a citation and/or potential source of further information does not
mean that the publisher and authors endorse the information or services the organization, website, or product may provide or
recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering pro-
fessional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with
a specialist where appropriate. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial
damages, including but not limited to special, incidental, consequential, or other damages. Further, readers should be aware
that websites listed in this work may have changed or disappeared between when this work was written and when it is read.
Library of Congress Cataloging-in-Publication Data
ISBN 978-1-119-76225-6
Cover images: Pixabay.Com

Cover design by Russell Richardson
Set in size of 11pt and Minion Pro by Manila Typesetting Company, Makati, Philippines
Printed in the USA
10 9 8 7 6 5 4 3 2 1
Contents
Preface xix
Part I: Conceptual Aspects on Cloud and Applications
of Machine Learning 1
1 Hybrid Cloud: A New Paradigm in Cloud Computing 3
Moumita Deb and Abantika Choudhury
1.1 Introduction 3
1.2 Hybrid Cloud 5
1.2.1 Architecture 6
1.2.2 Why Hybrid Cloud is Required? 6
1.2.3 Business and Hybrid Cloud 7
1.2.4 Things to Remember When Deploying Hybrid Cloud 8
1.3 Comparison Among Different Hybrid Cloud Providers 9
1.3.1 Cloud Storage and Backup Benefits 11
1.3.2 Pros and Cons of Different Service Providers 11
1.3.2.1 AWS Outpost 12
1.3.2.2 Microsoft Azure Stack 12
1.3.2.3 Google Cloud Anthos 12
1.3.3 Review on Storage of the Providers 13
1.3.3.1 AWS Outpost Storage 13
1.3.3.2 Google Cloud Anthos Storage 13
1.3.4 Pricing 15
1.4 Hybrid Cloud in Education 15
1.5 Significance of Hybrid Cloud Post-Pandemic 15
1.6 Security in Hybrid Cloud 16
1.6.1 Role of Human Error in Cloud Security 18
1.6.2 Handling Security Challenges 18
1.7 Use of AI in Hybrid Cloud 19
1.8 Future Research Direction 21
1.9 Conclusion 22
References 22
v
vi Contents
2 Recognition of Differentially Expressed Glycan Structure

of H1N1 Virus Using Unsupervised Learning Framework 25
Shillpi Mishrra
2.1 Introduction 25
2.2 Proposed Methodology 27
2.3 Result 28
2.3.1 Description of Datasets 29
2.3.2 Analysis of Result 29
2.3.3 Validation of Results 31
2.3.3.1 T-Test (Statistical Validation) 31
2.3.3.2 Statistical Validation 33
2.3.4 Glycan Cloud 37
2.4 Conclusions and Future Work 38
References 39
3 Selection of Certain Cancer Mediating Genes Using a Hybrid
Model Logistic Regression Supported by Principal Component
Analysis (PC-LR) 41
Subir Hazra, Alia Nikhat Khurshid and Akriti
3.1 Introduction 41
3.2 Related Methods 44
3.3 Methodology 46
3.3.1 Description 47
3.3.2 Flowchart 49
3.3.3 Algorithm 49
3.3.4 Interpretation of the Algorithm 50
3.3.5 Illustration 50
3.4 Result 51
3.4.1 Description of the Dataset 51
3.4.2 Result Analysis 51
3.4.3 Result Set Validation 52
3.5 Application in Cloud Domain 56
3.6 Conclusion 58
References 59
Part II: Cloud Security Systems Using Machine Learning

Techniques 61
4 Cost-Effective Voice-Controlled Real-Time Smart Informative
Interface Design With Google Assistance Technology 63
Soumen Santra, Partha Mukherjee and Arpan Deyasi
4.1 Introduction 64
4.2 Home Automation System 65
4.2.1 Sensors 65
4.2.2 Protocols 66
4.2.3 Technologies 66
Contents vii
4.2.4 Advantages 67
4.2.5 Disadvantages 67
4.3 Literature Review 67
4.4 Role of Sensors and Microcontrollers in Smart Home Design 68
4.5 Motivation of the Project 70
4.6 Smart Informative and Command Accepting Interface 70
4.7 Data Flow Diagram 71
4.8 Components of Informative Interface 72
4.9 Results 73
4.9.1 Circuit Design 73
4.9.2 LDR Data 76
4.9.3 API Data 76
4.10 Conclusion 78
4.11 Future Scope 78
References 78
5 Symmetric Key and Artificial Neural Network With Mealy Machine:
A Neoteric Model of Cryptosystem for Cloud Security 81
Anirban Bhowmik, Sunil Karforma and Joydeep Dey
5.1 Introduction 81
5.3 The Problem 86
5.4 Objectives and Contributions 86
5.5 Methodology 87
5.6 Results and Discussions 91
5.6.1 Statistical Analysis 93
5.6.2 Randomness Test of Key 94
5.6.3 Key Sensitivity Analysis 95
5.6.4 Security Analysis 96
5.6.5 Dataset Used on ANN 96
5.6.6 Comparisons 98
5.7 Conclusions 99
References 99
6 An Efficient Intrusion Detection System on Various Datasets Using
Machine Learning Techniques 103
Debraj Chatterjee
6.1 Introduction 103
6.2 Motivation and Justification of the Proposed Work 104
6.3 Terminology Related to IDS 105
6.3.1 Network 105
6.3.2 Network Traffic 105
6.3.3 Intrusion 106
6.3.4 Intrusion Detection System 106
6.3.4.1 Various Types of IDS 108
6.3.4.2 Working Methodology of IDS 108
viii Contents
6.3.4.3 Characteristics of IDS 109

6.3.4.4 Advantages of IDS 110
6.3.4.5 Disadvantages of IDS 111
6.3.5 Intrusion Prevention System (IPS) 111
6.3.5.1 Network-Based Intrusion Prevention System (NIPS) 111
6.3.5.2 Wireless Intrusion Prevention System (WIPS) 112
6.3.5.3 Network Behavior Analysis (NBA) 112
6.3.5.4 Host-Based Intrusion Prevention System (HIPS) 112
6.3.6 Comparison of IPS With IDS/Relation Between IDS and IPS 112
6.3.7 Different Methods of Evasion in Networks 113
6.4 Intrusion Attacks on Cloud Environment 114
6.5 Comparative Studies 116
6.6 Proposed Methodology 121
6.7 Result 122
6.8 Conclusion and Future Scope 125
References 126
7 You Are Known by Your Mood: A Text-Based Sentiment Analysis
for Cloud Security 129
Abhijit Roy and Parthajit Roy
7.3 Essential Prerequisites 133
7.3.1 Security Aspects 133
7.3.2 Machine Learning Tools 135
7.3.2.1 Naïve Bayes Classifier 135
7.3.2.2 Artificial Neural Network 136
7.4 Proposed Model 136
7.5 Experimental Setup 138
7.6 Results and Discussions 139
7.7 Application in Cloud Security 142
7.7.1 Ask an Intelligent Security Question 142
7.7.2 Homomorphic Data Storage 142
7.7.3 Information Diffusion 144
7.8 Conclusion and Future Scope 144
References 145
8 The State-of-the-Art in Zero-Knowledge Authentication Proof
for Cloud 149
Priyanka Ghosh
8.2 Attacks and Countermeasures 153
8.2.1 Malware and Ransomware Breaches 154
8.2.2 Prevention of Distributing Denial of Service 154
8.2.3 Threat Detection 154
8.3 Zero-Knowledge Proof 154
Contents ix
8.4 Machine Learning for Cloud Computing 156

8.4.1 Types of Learning Algorithms 156
8.4.1.1 Supervised Learning 156
8.4.1.2 Supervised Learning Approach 156
8.4.1.3 Unsupervised Learning 157
8.4.2 Application on Machine Learning for Cloud Computing 157
8.4.2.1 Image Recognition 157
8.4.2.2 Speech Recognition 157
8.4.2.3 Medical Diagnosis 158
8.4.2.4 Learning Associations 158
8.4.2.5 Classification 158
8.4.2.6 Prediction 158
8.4.2.7 Extraction 158
8.4.2.8 Regression 158
8.4.2.9 Financial Services 159
8.5 Zero-Knowledge Proof: Details 159
8.5.1 Comparative Study 159
8.5.1.1 Fiat-Shamir ZKP Protocol 159
8.5.2 Diffie-Hellman Key Exchange Algorithm 161
8.5.2.1 Discrete Logarithm Attack 161
8.5.2.2 Man-in-the-Middle Attack 162
8.5.3 ZKP Version 1 162
8.5.4 ZKP Version 2 162
8.5.5 Analysis 164
8.5.6 Cloud Security Architecture 166
8.5.7 Existing Cloud Computing Architectures 167
8.5.8 Issues With Current Clouds 167
8.6 Conclusion 168
References 169
9 A Robust Approach for Effective Spam Detection Using Supervised
Learning Techniques 171
Amartya Chakraborty, Suvendu Chattaraj, Sangita Karmakar
and Shillpi Mishrra
9.3 Motivation 174
9.4 System Overview 175
9.5 Data Description 176
9.6 Data Processing 176
9.7 Feature Extraction 178
9.8 Learning Techniques Used 179
9.8.1 Support Vector Machine 179
9.8.2 k-Nearest Neighbors 180
9.8.3 Decision Tree 180
9.8.4 Convolutional Neural Network 180
x Contents
9.9 Experimental Setup 182

9.10 Evaluation Metrics 183
9.11 Experimental Results 185
9.11.1 Observations in Comparison With State-of-the-Art 187
9.12 Application in Cloud Architecture 188
9.13 Conclusion 189
References 190
10 An Intelligent System for Securing Network From Intrusion Detection
and Prevention of Phishing Attack Using Machine Learning Approaches 193
Sumit Banik, Sagar Banik and Anupam Mukherjee
10.1.1 Types of Phishing 195
10.1.1.1 Spear Phishing 195
10.1.1.2 Whaling 195
10.1.1.3 Catphishing and Catfishing 195
10.1.1.4 Clone Phishing 196
10.1.1.5 Voice Phishing 196
10.1.2 Techniques of Phishing 196
10.1.2.1 Link Manipulation 196
10.1.2.2 Filter Evasion 196
10.1.2.3 Website Forgery 196
10.1.2.4 Covert Redirect 197
10.3 Materials and Methods 199
10.3.1 Dataset and Attributes 199
10.3.2 Proposed Methodology 199
10.3.2.1 Logistic Regression 202
10.3.2.2 Naïve Bayes 202
10.3.2.3 Support Vector Machine 203
10.3.2.4 Voting Classification 203
10.4 Result Analysis 204
10.4.1 Analysis of Different Parameters for ML Models 204
10.4.2 Predictive Outcome Analysis in Phishing URLs Dataset 205
10.4.3 Analysis of Performance Metrics 206
10.4.4 Statistical Analysis of Results 210
‌10.4.4.1 ANOVA: Two-Factor Without Replication 210
10.4.4.2 ANOVA: Single Factor 210
10.5 Conclusion 210
References 211
Part III: Cloud Security Analysis Using Machine Learning

Techniques 213
11 Cloud Security Using Honeypot Network and Blockchain: A Review 215
Smarta Sangui and Swarup Kr Ghosh
*

Contents xi
11.2 Cloud Computing Overview 216

11.2.1 Types of Cloud Computing Services 216
11.2.1.1 Software as a Service 216
11.2.1.2 Infrastructure as a Service 218
11.2.1.3 Platform as a Service 218
11.2.2 Deployment Models of Cloud Computing 218
11.2.2.1 Public Cloud 218
11.2.2.2 Private Cloud 218
11.2.2.3 Community Cloud 219
11.2.2.4 Hybrid Cloud 219
11.2.3 Security Concerns in Cloud Computing 219
11.2.3.1 Data Breaches 219
11.2.3.2 Insufficient Change Control and Misconfiguration 219
11.2.3.3 Lack of Strategy and Security Architecture 220
11.2.3.4 Insufficient Identity, Credential, Access,
and Key Management 220
11.2.3.5 Account Hijacking 220
11.2.3.6 Insider Threat 220
11.2.3.7 Insecure Interfaces and APIs 220
11.2.3.8 Weak Control Plane 221
11.3 Honeypot System 221
11.3.1 VM (Virtual Machine) as Honeypot in the Cloud 221
11.3.2 Attack Sensing and Analyzing Framework 222
11.3.3 A Fuzzy Technique Against Fingerprinting Attacks 223
11.3.4 Detecting and Classifying Malicious Access 224
11.3.5 A Bayesian Defense Model for Deceptive Attack 224
11.3.6 Strategic Game Model for DDoS Attacks in Smart Grid 226
11.4 Blockchain 227
11.4.1 Blockchain-Based Encrypted Cloud Storage 228
11.4.2 Cloud-Assisted EHR Sharing via Consortium Blockchain 229
11.4.3 Blockchain-Secured Cloud Storage 230
11.4.4 Blockchain and Edge Computing–Based Security Architecture 230
11.4.5 Data Provenance Architecture in Cloud Ecosystem
Using Blockchain 231
11.6 Comparative Analysis 233
11.7 Conclusion 233
References 234
12 Machine Learning–Based Security in Cloud Database—A Survey 239
Utsav Vora, Jayleena Mahato, Hrishav Dasgupta, Anand Kumar
and Swarup Kr Ghosh
12.2 Security Threats and Attacks 241
12.3 Dataset Description 244
12.3.1 NSL-KDD Dataset 244
12.3.2 UNSW-NB15 Dataset 244
xii Contents
12.4 Machine Learning for Cloud Security 245

12.4.1 Supervised Learning Techniques 245
12.4.1.1 Support Vector Machine 245
12.4.1.2 Artificial Neural Network 247
12.4.1.3 Deep Learning 249
12.4.1.4 Random Forest 250
12.4.2 Unsupervised Learning Techniques 251
12.4.2.1 K-Means Clustering 252
12.4.2.2 Fuzzy C-Means Clustering 253
12.4.2.3 Expectation-Maximization Clustering 253
12.4.2.4 Cuckoo Search With Particle Swarm
Optimization (PSO) 254
12.4.3 Hybrid Learning Techniques 256
12.4.3.1 HIDCC: Hybrid Intrusion Detection Approach
in Cloud Computing 256
12.4.3.2 Clustering-Based Hybrid Model in Deep Learning
Framework 257
12.4.3.3 K-Nearest Neighbor–Based Fuzzy C-Means
Mechanism 258
12.4.3.4 K-Means Clustering Using Support Vector Machine 260
12.4.3.5 K-Nearest Neighbor–Based Artificial Neural
Network Mechanism 260
12.4.3.6 Artificial Neural Network Fused With Support
Vector Machine 261
12.4.3.7 Particle Swarm Optimization–Based Probabilistic
Neural Network 261
12.5 Comparative Analysis 262
12.6 Conclusion 264
References 267
13 Machine Learning Adversarial Attacks: A Survey Beyond 271
Chandni Magoo and Puneet Garg
13.2 Adversarial Learning 272
13.2.1 Concept 272
13.3 Taxonomy of Adversarial Attacks 273
13.3.1 Attacks Based on Knowledge 273
13.3.1.1 Black Box Attack (Transferable Attack) 273
13.3.1.2 White Box Attack 274
13.3.2 Attacks Based on Goals 275
13.3.2.1 Target Attacks 275
13.3.2.2 Non-Target Attacks 275
13.3.3 Attacks Based on Strategies 275
13.3.3.1 Poisoning Attacks 275
13.3.3.2 Evasion Attacks 276
Contents xiii
13.3.4 Textual-Based Attacks (NLP) 276

13.3.4.1 Character Level Attacks 276
13.3.4.2 Word-Level Attacks 276
13.3.4.3 Sentence-Level Attacks 276
13.4 Review of Adversarial Attack Methods 276
13.4.1 L-BFGS 277
13.4.2 Feedforward Derivation Attack (Jacobian Attack) 277
13.4.3 Fast Gradient Sign Method 278
13.4.4 Methods of Different Text-Based Adversarial Attacks 278
13.4.5 Adversarial Attacks Methods Based on Language Models 284
13.4.6 Adversarial Attacks on Recommender Systems 284
13.4.6.1 Random Attack 284
13.4.6.2 Average Attack 286
13.4.6.3 Bandwagon Attack 286
13.4.6.4 Reverse Bandwagon Attack 286
13.5 Adversarial Attacks on Cloud-Based Platforms 287
13.6 Conclusion 288
References 288
14 Protocols for Cloud Security 293
Weijing You and Bo Chen
14.2 System and Adversarial Model 295
14.2.1 System Model 295
14.2.2 Adversarial Model 295
14.3 Protocols for Data Protection in Secure Cloud Computing 296
14.3.1 Homomorphic Encryption 297
14.3.2 Searchable Encryption 298
14.3.3 Attribute-Based Encryption 299
14.3.4 Secure Multi-Party Computation 300
14.4 Protocols for Data Protection in Secure Cloud Storage 301
14.4.1 Proofs of Encryption 301
14.4.2 Secure Message-Locked Encryption 303
14.4.3 Proofs of Storage 303
14.4.4 Proofs of Ownership 305
14.4.5 Proofs of Reliability 306
14.5 Protocols for Secure Cloud Systems 309
14.6 Protocols for Cloud Security in the Future 309
14.7 Conclusion 310
References 311
xiv Contents
Part IV: Case Studies Focused on Cloud Security 313

15 A Study on Google Cloud Platform (GCP) and Its Security 315
Agniswar Roy, Abhik Banerjee and Navneet Bhardwaj
15.1.1 Google Cloud Platform Current Market Holding 316
15.1.1.1 The Forrester Wave 317
15.1.1.2 Gartner Magic Quadrant 317
15.1.2 Google Cloud Platform Work Distribution 317
15.1.2.1 SaaS 318
15.1.2.2 PaaS 318
15.1.2.3 IaaS 318
15.1.2.4 On-Premise 318
15.2 Google Cloud Platform’s Security Features Basic Overview 318
15.2.1 Physical Premises Security 319
15.2.2 Hardware Security 319
15.2.3 Inter-Service Security 319
15.2.4 Data Security 320
15.2.5 Internet Security 320
15.2.6 In-Software Security 320
15.2.7 End User Access Security 321
15.3 Google Cloud Platform’s Architecture 321
15.3.1 Geographic Zone 321
15.3.2 Resource Management 322
15.3.2.1 IAM 322
15.3.2.2 Roles 323
15.3.2.3 Billing 323
15.4 Key Security Features 324
15.4.1 IAP 324
15.4.2 Compliance 325
15.4.3 Policy Analyzer 326
15.4.4 Security Command Center 326
15.4.4.1 Standard Tier 326
15.4.4.2 Premium Tier 326
15.4.5 Data Loss Protection 329
15.4.6 Key Management 329
15.4.7 Secret Manager 330
15.4.8 Monitoring 330
15.5 Key Application Features 330
15.5.1 Stackdriver (Currently Operations) 330
15.5.1.1 Profiler 330
15.5.1.2 Cloud Debugger 330
15.5.1.3 Trace 331
15.5.2 Network 331
15.5.3 Virtual Machine Specifications 332
Contents xv
15.5.4 Preemptible VMs 332

15.6 Computation in Google Cloud Platform 332
15.6.1 Compute Engine 332
15.6.2 App Engine 333
15.6.3 Container Engine 333
15.6.4 Cloud Functions 333
15.7 Storage in Google Cloud Platform 333
15.8 Network in Google Cloud Platform 334
15.9 Data in Google Cloud Platform 334
15.10 Machine Learning in Google Cloud Platform 335
15.11 Conclusion 335
References 337
16 Case Study of Azure and Azure Security Practices 339
Navneet Bhardwaj, Abhik Banerjee and Agniswar Roy
16.1.1 Azure Current Market Holding 340
16.1.2 The Forrester Wave 340
16.1.3 Gartner Magic Quadrant 340
16.2 Microsoft Azure—The Security Infrastructure 341
16.2.1 Azure Security Features and Tools 341
16.2.2 Network Security 342
16.3 Data Encryption 342
16.3.1 Data Encryption at Rest 342
16.3.2 Data Encryption at Transit 342
16.3.3 Asset and Inventory Management 343
16.3.4 Azure Marketplace 343
16.4 Azure Cloud Security Architecture 344
16.4.1 Working 344
16.4.2 Design Principles 344
16.4.2.1 Alignment of Security Policies 344
16.4.2.2 Building a Comprehensive Strategy 345
16.4.2.3 Simplicity Driven 345
16.4.2.4 Leveraging Native Controls 345
16.4.2.5 Identification-Based Authentication 345
16.4.2.6 Accountability 345
16.4.2.7 Embracing Automation 345
16.4.2.8 Stress on Information Protection 345
16.4.2.9 Continuous Evaluation 346
16.4.2.10 Skilled Workforce 346
16.5 Azure Architecture 346
16.5.1 Components 346
16.5.1.1 Azure Api Gateway 346
16.5.1.2 Azure Functions 346
16.5.2 Services 347
16.5.2.1 Azure Virtual Machine 347
xvi Contents
16.5.2.2 Blob Storage 347

16.5.2.3 Azure Virtual Network 348
16.5.2.4 Content Delivery Network 348
16.5.2.5 Azure SQL Database 349
16.6 Features of Azure 350
16.6.1 Key Features 350
16.6.1.1 Data Resiliency 350
16.6.1.2 Data Security 350
16.6.1.3 BCDR Integration 350
16.6.1.4 Storage Management 351
16.6.1.5 Single Pane View 351
16.7 Common Azure Security Features 351
16.7.1 Security Center 351
16.7.2 Key Vault 351
16.7.3 Azure Active Directory 352
16.7.3.1 Application Management 352
16.7.3.2 Conditional Access 352
16.7.3.3 Device Identity Management 352
16.7.3.4 Identity Protection 353
16.7.3.5 Azure Sentinel 353
16.7.3.6 Privileged Identity Management 354
16.7.3.7 Multifactor Authentication 354
16.7.3.8 Single Sign On 354
16.8 Conclusion 355
References 355
17 Nutanix Hybrid Cloud From Security Perspective 357
Abhik Banerjee, Agniswar Roy, Amar Kalvikatte and Navneet Bhardwaj
17.2 Growth of Nutanix 358
17.2.1 Gartner Magic Quadrant 358
17.2.2 The Forrester Wave 358
17.2.3 Consumer Acquisition 359
17.2.4 Revenue 359
17.3 Introductory Concepts 361
17.3.1 Plane Concepts 361
17.3.1.1 Control Plane 361
17.3.1.2 Data Plane 361
17.3.2 Security Technical Implementation Guides 362
17.3.3 SaltStack and SCMA 362
17.4 Nutanix Hybrid Cloud 362
17.4.1 Prism 362
17.4.1.1 Prism Element 363
17.4.1.2 Prism Central 364
17.4.2 Acropolis 365
17.4.2.1 Distributed Storage Fabric 365
Contents xvii
17.4.2.2 AHV 367

17.5 Reinforcing AHV and Controller VM 367
17.6 Disaster Management and Recovery 368
17.6.1 Protection Domains and Consistent Groups 368
17.6.2 Nutanix DSF Replication of OpLog 369
17.6.3 DSF Snapshots and VmQueisced Snapshot Service 370
17.6.4 Nutanix Cerebro 370
17.7 Security and Policy Management on Nutanix Hybrid Cloud 371
17.7.1 Authentication on Nutanix 372
17.7.2 Nutanix Data Encryption 372
17.7.3 Security Policy Management 373
17.7.3.1 Enforcing a Policy 374
17.7.3.2 Priority of a Policy 374
17.7.3.3 Automated Enforcement 374
17.8 Network Security and Log Management 374
17.8.1 Segmented and Unsegmented Network 375
17.9 Conclusion 376
References 376
Part V: Policy Aspects 379

18 A Data Science Approach Based on User Interactions to Generate
Access Control Policies for Large Collections of Documents 381
Jedidiah Yanez-Sierra, Arturo Diaz-Perez and Victor Sosa-Sosa
18.2 Related Work 383
18.3 Network Science Theory 384
18.4 Approach to Spread Policies Using Networks Science 387
18.4.1 Finding the Most Relevant Spreaders 388
18.4.1.1 Weighting Users 389
18.4.1.2 Selecting the Top  Spreaders 390
18.4.2 Assign and Spread the Access Control Policies 390
18.4.2.1 Access Control Policies 391
18.4.2.2 Horizontal Spreading 391
18.4.2.3 Vertical Spreading (Bottom-Up) 392
18.4.2.4 Policies Refinement 395
18.4.3 Structural Complexity Analysis of CP-ABE Policies 395
18.4.3.1 Assessing the WSC for ABE Policies 396
18.4.3.2 Assessing the Policies Generated in the Spreading
Process 397
18.4.4 Effectiveness Analysis 398
18.4.4.1 Evaluation Metrics 399
18.4.4.2 Adjusting the Interaction Graph to Assess Policy
Effectiveness 400
18.4.4.3 Method to Complement the User Interactions
(Synthetic Edges Generation) 400
xviii Contents
18.4.5 Measuring Policy Effectiveness in the User Interaction Graph 403

18.4.5.1 Simple Node-Based Strategy 403
18.4.5.2 Weighted Node-Based Strategy 404
18.5 Evaluation 405
18.5.1 Dataset Description 405
18.5.2 Results of the Complexity Evaluation 406
18.5.3 Effectiveness Results From the Real Edges 407
18.5.4 Effectiveness Results Using Real and Synthetic Edges 408
18.5.4.1 Results of the Effectiveness Metrics for the Enhanced
G+ Graph 410
18.6 Conclusions 413
References 414
19 AI, ML, & Robotics in iSchools: An Academic Analysis for an Intelligent
Societal Systems 417
P. K. Paul
19.2 Objective 419
19.3 Methodology 420
19.3.1 iSchools, Technologies, and Artificial Intelligence,
ML, and Robotics 420
19.4 Artificial Intelligence, ML, and Robotics: An Overview 427
19.5 Artificial Intelligence, ML, and Robotics as an Academic Program:
A Case on iSchools—North American Region 428
19.6 Suggestions 431
19.7 Motivation and Future Works 435
19.8 Conclusion 435
References 436
Index 439
Preface
Our objective in writing this book was to provide the reader with an in-depth knowledge
of how to integrate machine learning (ML) approaches to meet various analytical issues
in cloud security deemed necessary due to the advancement of IoT networks. Although
one of the ways to achieve cloud security is by using ML, the technique has long-standing
challenges that require methodological and theoretical approaches. Therefore, because the
conventional cryptographic approach is less frequently applied in resource-constrained
devices, the ML approach may be effectively used in providing security in the constantly
growing cloud environment. Machine learning algorithms can also be used to meet various
cloud security issues for effective intrusion detection and zero-knowledge authentication
systems. Moreover, these algorithms can also be used in applications and for much more,
including measuring passive attacks and designing protocols and privacy systems. This
book contains case studies/projects for implementing some security features based on ML
algorithms and analytics. It will provide learning paradigms for the field of artificial intelli-
gence and the deep learning community, with related datasets to help delve deeper into ML
for cloud security.
This book is organized into five parts. As the entire book is based on ML techniques,
the three chapters contained in “Part I: Conceptual Aspects of Cloud and Applications of
Machine Learning,” describe cloud environments and ML methods and techniques. The
seven chapters in “Part II: Cloud Security Systems Using Machine Learning Techniques,”
describe ML algorithms and techniques which are hard coded and implemented for pro-
viding various security aspects of cloud environments. The four chapters of “Part III: Cloud
Security Analysis Using Machine Learning Techniques,” present some of the recent studies
and surveys of ML techniques and analytics for providing cloud security. The next three
chapters in “Part IV: Case Studies Focused on Cloud Security,” are unique to this book as
they contain three case studies of three cloud products from a security perspective. These
three products are mainly in the domains of public cloud, private cloud and hybrid cloud.
Finally, the two chapters in “Part V: Policy Aspects,” pertain to policy aspects related to
the cloud environment and cloud security using ML techniques and analytics. Each of the
chapters mentioned above are individually highlighted chapter by chapter below.
Part I: Conceptual Aspects of Cloud and Applications of Machine Learning

–– Chapter 1 begins with an introduction to various parameters of cloud such
as scalability, cost, speed, reliability, performance and security. Next, hybrid
cloud is discussed in detail along with cloud architecture and how it func-
tions. A brief comparison of various cloud providers is given next. After the
xix
xx Preface
use of cloud in education, finance, etc., is described, the chapter concludes

with a discussion of security aspects of a cloud environment.
–– Chapter 2 discusses how to recognize differentially expressed glycan struc-
ture of H1N1 virus using unsupervised learning framework. This chap-
ter gives the reader a better understanding of machine learning (ML) and
analytics. Next, the detailed workings of an ML methodology are presented
along with a flowchart. The result part of this chapter contains the analytics
for the ML technique.
–– Chapter 3 presents a hybrid model of logistic regression supported by PC-LR
to select cancer mediating genes. This is another good chapter to help bet-
ter understand ML techniques and analytics. It provides the details of an
ML learning methodology and algorithms with results and analysis using
datasets.
Part II: Cloud Security Systems Using Machine Learning Techniques

–– Chapter 4 shows the implementation of a voice-controlled real-time smart
informative interface design with Google assistance technology that is more
cost-effective than the existing products on the market. This system can be
used for various cloud-based applications such as home automation. It uses
microcontrollers and sensors in smart home design which can be connected
through cloud database. Security concerns are also discussed in this chapter.
–– Chapter 5 discusses a neoteric model of a cryptosystem for cloud security
by using symmetric key and artificial neural network with Mealy machine.
A cryptosystem is used to provide data or information confidentiality and a
state-based cryptosystem is implemented using Mealy machine. This chapter
gives a detailed algorithm with results generated using Lenovo G80 with pro-
cessor Intel® Pentium® CPU B950@210GHz and RAM 2GB and program-
ming language Turbo C, DebC++ and disc drive SA 9500326AS ATA and
Windows 7 Ultimate (32 Bits) OS.
–– Chapter 6 describes the implementation of an effective intrusion detection
system using ML techniques through various datasets. The chapter begins
with a description of an intrusion detection system and how it is beneficial
for cloud environment. Next, various intrusion attacks on cloud environment
are described along with a comparative study. Finally, a proposed methodol-
ogy of IDS in cloud environment is given along with implementation results.
–– Chapter 7 beautifully describes text-based sentiment analysis for cloud secu-
rity that extracts the mood of users in a cloud environment, which is an
evolving topic in ML. A proposed model for text-based sentiment analysis
is presented along with an experimental setup with implementation results.
Since text-based sentiment analysis potentially identifies malicious users in a
cloud environment, the chapter concludes with applications of this method
and implementation for cloud security.
–– Chapter 8 discusses zero-knowledge proof (ZKP) for cloud, which is a
method for identifying legitimate users without revealing their identity. The
ZKP consist of three parts: the first is ticket generator, the second is user,
Preface xxi
and the third is verifier. For example, to see a movie in a theater we purchase
ticket. So, the theater counter is the ticket generator; and while purchasing
a ticket here we generally don’t reveal our identifying information such as
name, address or social security number. We are allowed to enter the theater
when this ticket is verified at the gate, so, this is the verifier algorithm. This
chapter also discusses ZKP for cloud security.
–– Chapter 9 discusses an effective spam detection system for cloud secu-
rity using supervised ML techniques. Spam, which is an unwanted mes-
sage that contains malicious links, viral attachments, unwelcome images
and misinformation, is a major security concern for any digital system and
requires an effective spam detection system. Therefore, this chapter begins
by discussing the requirements for such a system. Then, it gradually moves
towards a supervised ML-technique-based spam detection system, mainly
using a support vector machine (SVM) and convolutional neural network
(CNN). Implementation results are also given with application in cloud
environment.
–– Chapter 10 describes an intelligent system for securing network from intru-
sion detection and phishing attacks using ML approaches, with a focus on
phishing attacks on the cloud environment. It begins by describing different
fishing attacks on cloud environment and then proposes a method for detect-
ing these attacks using ML. Next, analysis of different parameters for ML
models, predictive outcome analysis in phishing URLs dataset, analysis of
performance metrics and statistical analysis of results are presented.
Part III: Cloud Security Analysis Using Machine Learning Techniques

–– Chapter 11 discusses cloud security using honeypot network and blockchain.
It begins with an overview of cloud computing and then describes cloud
computing deployment models and security concerns in cloud computing.
Then the honeypot network and its system design are discussed, followed by
the use of blockchain-based honeypot network. A good comparative analysis
is given at the end of the chapter.
–– Chapter 12 includes a survey on ML-based security in cloud database. The
chapter starts with a discussion of the various ML techniques used to provide
security in a cloud database. Then a study is presented which mainly con-
sists of three parts: first, supervised learning methods, such as support vector
machine (SVM), artificial neural network, etc., are given; second, unsuper-
vised learning methods, such as K-means clustering, fuzzy C-means cluster-
ing, etc., are given; third, hybrid learning techniques, such as hybrid intrusion
detection approach (HIDCC) in cloud computing, clustering-based hybrid
model in deep learning framework, etc., are given. Comparative analyses are
also given at the end.
–– Chapter 13 provides a survey on ML-based adversarial attacks on cloud
environment. The chapter starts with the concepts of adversarial learning
followed by the taxonomy of adversarial attacks. Various algorithms found
in the literature for ML-based adversarial attacks on cloud environment are
xxii Preface
also presented. Then, various studies on adversarial attacks on cloud-based

platforms and their comparative studies are discussed.
–– Chapter 14 provides a detailed study of the protocols used for cloud secu-
rity. The chapter starts by discussing the system and adversarial models, and
then the protocols for data protection in secure cloud computing are given
followed by a discussion of the protocols for data protection in secure cloud
storage. Finally, various protocols for secure cloud systems are discussed. The
authors also attempt to give a futuristic view of the protocols that may be
implemented for cloud security.
Part IV: Case Studies Focused on Cloud Security

–– Chapter 15 is a detailed presentation of the Google cloud platform (GCP) and
its security features. It begins by discussing GCP’s current market holdings
and then describes the work distribution in GCP. Next, the chapter gradually
moves towards a basic overview of security features in GCP and describes the
GCP architecture along with its key security and application features. Then,
an interesting part is presented that describes various computations used in
GCP, followed by a discussion of the storage, network, data and ML policies
used in GCP.
–– Chapter 16 presents a case study of Microsoft Azure cloud and its security
features. The beginning of the chapter covers Azure’s current market hold-
ings and the Forrester Wave and Gartner Magic Quadrant reports. Then, the
security infrastructure of Azure is given, which covers its security features
and tools, Azure network security, data encryption used in Azure, asset and
inventory management, and the Azure marketplace. Next, details of Azure
cloud security architecture are presented along with its working and design
principles, followed by the components and services of Azure architecture.
The chapter ends with a discussion of its various features and why Azure is
gaining popularity.
–– Chapter 17 presents a case study on Nutanix hybrid cloud from a security
perspective. Nutanix is a fast-growing hybrid cloud in the current scenario.
The chapter begins with the growth of Nutanix and then presents introduc-
tory concepts about it. Next, Nutanix hybrid cloud architecture is discussed
in relation to computation, storage and networking. Then, reinforcing AHV
and controller VM are described, followed by disaster management and
recovery used in Nutanix hybrid cloud. A detailed study on security and pol-
icy management in Nutanix hybrid cloud is then presented. The chapter con-
cludes with a discussion of network security and log management in Nutanix
hybrid cloud.
Part V: Policy Aspects

–– Chapter 18 describes a data science approach based on user interactions to
generate access control policies for large collections of documents in cloud
environment. After a general introduction to network science theory, various
Preface xxiii
approaches for spreading policies using network science are discussed. Then,
evaluations and matrices to evaluate policies for cloud security are described.
This chapter concludes with a presentation of all the simulation results.
–– Chapter 19 discusses the policies of iSchools with artificial intelligence,
machine learning, and robotics through analysis of programs, curriculum and
potentialities towards intelligent societal systems on cloud platform. iSchools
are a kind of consortium that develops with the collection of information and
technology-related schools and academic units. In the last decade there has
been a significant growth in the development of such academic bodies. This
chapter provides a policy framework for iSchools, the methodology involved
and a list of available iSchools. The chapter concludes with some policy sug-
gestions and future work related to iSchools.
The Editors
October 2021
Part I
CONCEPTUAL ASPECTS ON CLOUD AND
APPLICATIONS OF MACHINE LEARNING
1
Hybrid Cloud: A New Paradigm
in Cloud Computing
Moumita Deb* and Abantika Choudhury†
RCC Institute of Information Technology, Kolkata, West Bengal, India
Abstract
Hybrid cloud computing is basically a combination of cloud computing with on-premise resources
to provide work portability, load distribution, and security. Hybrid cloud may include one public
and one private cloud, or it may contain two or more private clouds or may have two or more public
clouds depending on the requirement. Public clouds are generally provided by third party vendors
like Amazon, Google, and Microsoft. These clouds traditionally ran off premise and provide ser-
vices through internet. Whereas private clouds also offer computing services to selected user either
over the internet or within a private internal network and conventionally ran on-premise. But this
scenario is changing nowadays. Earlier distinction between private and public clouds can be done
on the location and ownership information, but currently, public clouds are running in on-premise
data centers of customer and private clouds are constructed on off premise rented, vendor-owned
data centers as well. So, the architecture is becoming complex. Hybrid cloud reduces the potential
exposure of sensitive or crucial data from the public while keeping non-sensitive data into the cloud.
Thus, secure access to data while enjoying attractive services of the public cloud is the key factor in
hybrid cloud. Here, we have done a survey on hybrid cloud as it is one of the most promising areas
in cloud computing, discuss all insight details. Security issues and measures in hybrid cloud are also
discussed along with the use of artificial intelligence. We do not intend to propose any new findings
rather we will figure out some of future research directions.
Keywords: PaaS, SaaS, IaaS, SLA, agility, encryption, middleware, AI
1.1 Introduction
Cloud computing is catering computing services such as storage, networking, servers, ana-
lytics, intelligence, and software though the internet on demand basis. We typically have
to pay for only for the services we use. IT is a growing industry and catering its service
requirement is challenging. On-premise resources are not sufficient always, so leveraging
attractive facilities provided by cloud service providers is often required. Typical services
*Corresponding author: moudeb@gmail.com

†Corresponding author: abantika_choudhury@rediffmail.com
Rajdeep Chakraborty, Anupam Ghosh and Jyotsna Kumar Mandal (eds.) Machine Learning Techniques and Analytics for
Cloud Security, (3–24) © 2022 Scrivener Publishing LLC
3
4 Machine Learning Techniques and Analytics for Cloud Security
provided by cloud computing are Platform as a service (PaaS), Software as a service (SaaS),
and Infrastructure as a service (IaaS). But all the clouds are not same and no one particular
cloud can satisfy all the customer. As a result, various types of services are emerging to cater
the need of any organization. The following are the facilities cater by cloud computing.
• Scalability: IT services are not restricted to offline resources anymore,

online cloud services can do a wonder. Any business can be extended
based on the market need through the use of cloud computing services. A
client needs almost nothing but a computer with internet connection, rest
of the services can be borrowed from cloud vendors. Business can grow
according to the requirement. Scalability is the key factor in adoption of
any new paradigm. An organization meant for 100 people can be easily
scaled up to 1,000 (ideally any number) people with the help of the cloud
computing services.
• Cost: Since cloud provides services pay as you use basis, cost of setting up a
business has reduced manifolds. Capital expense in buying server, software,
and experts for managing infrastructure is not mandatory anymore; vendors
can provide all these services. Cost saving is one of the most lucrative features
of cloud computing. Any startup company can afford the cost of the setup
price required for the orchestration of public cloud; thus, they can engage
their selves exclusively for the development of their business.
• Speed: Cloud computing helps to speed up the overall functioning of any
organization. Several lucrative easy-to-use options are just one click away, so
designers and programmers can freely think about their innovations, and as a
result, the speed and performance can be enhanced. Moreover, since most of
the background hazards are handled by the cloud service providers as a result
implementation of any advanced thinking can be made possible quickly and
effortlessly.
• Reliability: Reliability is a key factor where huge data need to handle all the
time. Periodic data backup and use of disaster recovery methods helps to
increase the data reliability in cloud computing. Also, since space is not a
constraint anymore, clients can keep mirrored data. A reliable system often
leads to a secure system. Any organizations need to handle huge user centric
sensitive data as well as business related data. Maintaining the reliability in
the data need several rules and regulations to be enforced.
• Performance: Improved operation, better customer support, and flexible
workplace aid companies to perform better than conventional on-premise
system. Amazon helps Car company Toyota to build cloud-based data cen-
ters. The company is going to use the behavioral data of the user of the car,
and based on that, they will send service and insurance related data [1]. User
can also use Facebook or Twitter in their car dashboard. This is only an exam-
ple; there is lot more. Adaptation of advanced technology excels the perfor-
mance of existing system as cloud plays a crucial role here.
Hybrid Cloud: New Paradigm in Cloud Computing 5
• Security: Cloud service providers use many security mechanisms like

encryption, authentication of user, authorization, and use of some Artificial
Intelligence (AI)–based method to secure their app, data, and infrastructure
from possible threats.
A combination of secure open source technologies along with integrated network may
be used for secure hybrid cloud deployment like it does in HCDM [16]. But, before deploy-
ment, the customer need to determine what type of cloud computing architecture is best
suitable. There are three different ways to organize cloud: private, public, and hybrid. Here,
we will discuss about hybrid cloud, its benefits, and security aspects.
Thus, motivation of this review is to provide a broad details of hybrid cloud computing,
why it is gaining popularity, how business is going to be affected through the use of cloud
adaption in near future, what security aspects need to dealt by vendors, and how AI can
help in this regard. The following sections deal with all this topics.
1.2 Hybrid Cloud

If we go by the definition of National Institute of Standards and Technology [3], hybrid
cloud is a “composition of two or more different types of cloud infrastructure that are bind
together with the help of proprietary and standardized technology for the purpose of data
and application portability. So, Simple amalgamation of cloud and on-premise data should
not misinterpret as hybrid cloud. It should also provide the following facilities [2]:
• Workload distribution by portability.

• Networking between system and devices, by the use of LAN, WAN, or VPN.
• Use of a comprehensive unified automation tool.
• A complex powerful middleware for abstracting the background details.
• Incorporating availability and scalability of resources.
• Integrating disaster management and recovery strategies.
Thus, it enables the customer to extend their business by leveraging the attractive services
provided by public cloud as well as securing the delicate data through the use of private
cloud. When the demand of a business fluctuates that may be sudden peak in the business
come or sudden fall down, in those scenarios, hybrid cloud is the best possible option as
it has that flexibility [8]. Organizations can seamlessly use public cloud amenities without
directly giving access to their data centers which are part of their on-premise servers. So,
business critical data and applications can be kept safe behind, while computing power of
the public cloud can be used for doing complex tasks. Organizations will only have to pay
for the services it is using without considering the capital expenditure involve in purchasing,
programming and maintaining new resources which can be used for a short span of time and
may remain idle for long. Private cloud on the other hand is more like public cloud, but gen-
erally installed on clients datacenter and mainly focus on self-servicing, scalable structure.
Single tone service nature, service-level agreement (SLA), and similar association make the
relationship between client and cloud stronger and less demanding [33, 34].
1.2.1 Architecture
There may be any combination of cloud services when to deploy a hybrid cloud. It may
the client has its own on-premise private cloud as IaaS and leverage public cloud as SaaS.
Private cloud may be on premise or sometimes off premise on a dedicated server [10]. There
is no fixed fits for all architecture. Private clouds can be made individually, whereas public
cloud can be hired from vendors like Amazon, Microsoft, Alibaba, Google, and IBM. Next,
a middleware is required to combine public and private cloud mostly provided by the cloud
vendors as a part of their package. Figure 1.1 gives general diagram of a hybrid cloud.
In case of hybrid cloud architecture, the following is a list of properties that must to be
kept in mind [4]:
a. Multiple devices need to be connected via LAN, WAN, or VPN with a com-
mon middleware that provides an API for user services. Rather than using a
vast network of API, a single operating system must be used throughout the
network and APIs can be built on top of that.
b. Resources are made available to all the connected devices via virtualization
and it can be scaled up to any limit.
c. The middleware does all the coordination between devices and resources are
made available on demand basis with proper authentication.
1.2.2 Why Hybrid Cloud is Required?

Hybrid cloud means different service to different people [5]. Need of an organization
depends on diverse aspects of IT. As the perspective of application designer, business devel-
oper, and infrastructure support personnel is different from one another, their expectation
from the system also varies.
HYBRID CLOUD MODEL
On-Premise Apps
PUBLIC CLOUD
SQL SQL
Off-Premise Apps
SaaS, Iaas and PasS
Mobile Applications
PRIVATE CLOUD
Figure 1.1 General architecture of hybrid cloud.

• Application programmer always requires support for edge technologies.

Availability of high-end resources and cutting edge technology support is
the primary concern of a developer. Off premise support for such is essential.
Flexibility in deployment of changing technology services, speedy availability
of the new resources required by the solution, peak support for on-premise
system, and seamless and continuous integration of system services are key
issues need to be dealt in hybrid cloud. Disaster management is also an inte-
gral part of it.
• On the contrary, infrastructure support personnel always look for a steady
build in support for smooth execution of overall activities of the organiza-
tion. Off premise support for virtualized computing resources is often nec-
essary in IT. In this scenario, the role of infrastructure support team is very
crucial. Visibility of all the resources wherever it is, monitoring them in fed-
erated way following SLA, management of deployed setup for auditing and
security management, accessibility of all resources, and control provisioning
are key consideration in case of hybrid cloud.
• Business developer, on the other hand, focuses on consumer marketing in
cost-effective manner [6]. The need of IT business has manifolds. Support
for newly growing technology like mobile or web-based application requires
agile and easy to extend network, and at the same time, consistent system and
stable process management services cannot be replaced. So, business devel-
opers have to look into all these aspects, and at the same time, they have to
focus on the cost. The maintenance and management cost should not exceed
the overall financial budget. Looking at the SLAs and software license expo-
sure, they need to design financial plans that can fulfill the whole organiza-
tion’s prerequisites.
No matter how well we plan the future, it still remains uncertain and hybrid cloud pro-
vides the facility to use cloud services as and when it is required. It is also quite unlikely
that workload of an organization remains same throughout the whole year. Suppose an
organization is working on big data analytics, it can take help of public cloud computing
resources for high complex computations but that too is not needed for long run, may be
require for few months. Here, public cloud resources can be borrowed for few months only.
In the same way, startup companies can start with some trivial private resources and take
cloud services for rest of the processing. Then, based on the performance, they can plan
to expand the business with the help of public cloud. All these are possible only in case of
hybrid cloud as it has agility, scalability, data reliability, speedy recovery, and improved
connectivity and security.
1.2.3 Business and Hybrid Cloud

According to Hybrid Cloud Market report, in 2018, hybrid cloud market was USD 44.6 bil-
lion and expected to grow to USD 97.6 billion by the end of 2023 with Compound Annual
Growth Rate (CAGR) of 17.0% [9]. IaaS is expected to hold a large market in the fore-
cast period as it facilitates to migrate workload from on premise to off premise in high
peak hours. Hybrid web hosting also hold a big market as it provides management of all
Hybrid Cloud Market - Growth Rate by Region (2020 - 2025)
Regional Growth Rates

High
Mid
Low
Source: Mordor Intelligence
Figure 1.2 Market trend of hybrid cloud [14].
hosting services in just single point of contact. North America was the most promising
hybrid cloud market place in 2018 and Asia Pacific areas shows the highest CAGR. So,
hybrid cloud is a promising area in business. Major sectors using hybrid cloud computing
are healthcare, retail, government, or public sectors, banking, entertainment media, insur-
ance, finance, communication media, etc. [14]. According to a report published by Mordor
Intelligence, North America, Middle East, Africa, Europe, and Asia Pacific are top growing
regions worldwide. Figure 1.2 shows the hybrid cloud market. Green portions represent
highly growing market. Hybrid cloud management software solution is the main reason of
this popularity. Starting from deployment to quota management, customization of service
library, costing, performance management, and governance, everything is taken care of,
like the software management tool. Mostly, the services provided by public providers are
restricted to some architecture or technology and vendor specific. But the management
tool provided by hybrid providers helps to amalgamate different services provided by var-
ious vendors. Amazon and Microsoft, the giants in this field, are working hard in the up
gradation of their management software by including advanced infrastructure templates,
libraries, API, and apps. In India, IBM is also approaching toward hybrid cloud and AI [15].
IBM invested $1 billion into a cloud ecosystem project in the month of August. They are
expected to invest more in the coming time. In India, 17% of organizations are planning to
spend investment from 42% to 49% on hybrid cloud by 2023 according to a study by IBM
IBV. Since India is heading toward a digital transformation and self-reliant camping, so the
opportunity of new technology adaptation also increasing.
1.2.4 Things to Remember When Deploying Hybrid Cloud

Having an understanding what hybrid cloud is and how it facilitates the activities of any
organization, now, we need to understand some factors that have to be considered before
the deployment of hybrid cloud.
• Selection of best suitable platform for cloud: As discussed, the need of every
organization is not same. Before deployment of the hybrid cloud, organi-
zations need to have a plan for the services; it will borrow from the public
cloud. If it is going to use only SaaS, then it is not a problem but it is going
to use IaaS or PaaS and then it is very important to take the correct decision
from the commencement of the service as building a hybrid structure that
would not be able to handle additional workload generates severe problem.
• Whether to use unified OS or not: In true hybrid cloud, a unified OS is
installed in the middleware that basically governs the overall functionalities.
But in some cases, on-premise system may be operated by its own OS then
just with the help of internet they can connect to public cloud. The perfor-
mance of this architecture will be vast different from unified OS. OpenStack,
VMWare cloud, Nutanix, and Kubernetes are some example of cloud OS
framework. These frameworks are sufficient building the middleware and it
provides OS and all supporting application for the smooth execution of all
activities in hybrid cloud.
• How to manage different activity: Huge amount of data need to be handled
in case of hybrid cloud. A hybrid system should look into smooth accessi-
bility of data, and at the same time, security of data needs to be guaranteed.
Anyone cannot host any data onto the public cloud. Proper personnel with
adequate experience need to be engaged for the management of dedicated
applications.
• How security of data will be guaranteed: Since data is moving in between
public and private cloud, it needs to be secured. Through security mecha-
nisms of public cloud, it has developed much from its early date but still it is
not 100% secure. There are always threats of data breach. Migration of sensi-
tive need special care as sight alteration in business sensitive data might cause
severe problem in the business.
• How to integrate public cloud with existing on-premise system: Amalgamation
of public cloud onto an existing on-premise system often needs several alter-
ations in the working of the existing on-premise system. Overall performance
of the system should always improve with the addition of the public cloud,
and it should not degrade.
• How to manage common backup and disaster recovery: Data need to be
backed up to ensure reliability and availability. Backing up of all the data both
in private and public cloud need to be done. At the same time, the system
should be able to handle catastrophic failure or disaster. How to maintain a
common routine for all the operational data to accommodate those situa-
tions is key to the success of hybrid cloud deployment.
Building a hybrid cloud is a complex procedure but successful implementation will pro-
vide scalability, flexibility, security, and cost saving. More and more organizations approach-
ing toward hybrid cloud for the current benefit and future growth.
1.3 Comparison Among Different Hybrid Cloud Providers

The major famous leading cloud computing vendors are Google Cloud, AWS, and
Microsoft Azure. They have their some advantages and disadvantages. These three leading
cloud providers have important role in the PaaS and IaaS markets. Synergy Research Group
reported that the growth of Amazon is very significantly high in overall growth of market.
It possesses a share of 33% of cloud market throughout the world. In second position, there
is Microsoft. Microsoft is very fast growing and in the last four quarters, and its share has
been increased by 3% and it reaches at 18%. Nowadays, cloud computing is become much
matured. It is becoming hybrid cloud, and it also becomes more enhanced as market share.
New trends have come to improve cloud computing system in 2020 than that of 2017, 2018,
and 2019 [17].
Hybrid cloud [17] provides strategy for enterprises that involve operational part of vari-
eties of job in varieties infrastructure, whether on private cloud and public cloud with a
proprietary different layers at the top level. Multi-cloud concept is similar kind of but not
to involve any private cloud. Hybrid cloud is the most popular strategy among enterprises;
58% of respondents stated that it is their choice able approach while 10% for a single public
cloud provider and 17% for multiple public clouds.
• Microsoft Azure Stack: Microsoft is a popular vendor that provides

hybrid cloud. Because it has huge on-premises legacy. The services of MS
Azure are developed on Windows Server. The .Net framework and the
Visual Studio provide better features of on apps for their smoother ser-
vices [17].
• AWS Outposts: Amazon’s Amazon Web Services (AWS) is a one of the best
product. It is one of the most popular in market and its share is next to the
Microsoft leading competitor. This company has variety of services and larg-
est data center that continues to provide facilities to billions of customers.
AWS is very well-known public cloud that offers many services to connect
for installations to the cloud. It also serves everything like disaster recovery
and burst capacity [17].
• Google Cloud Anthos: The Google Cloud Platform is another one popular in
hybrid cloud. It is a competitor of Microsoft AWS and IBM. Google primarily
made pure cloud system, but later, they changed policy and started to work
with on-premise systems for disaster recovery, elastic infrastructure, Big
Data, and DevOps. It also provides a huge number of cloud-based services.
The services are based on AI efforts based on AI processor and TensorFlow.
No one can buy TensorFlow system but can run AI and machine learning
apps on Google Cloud [17, 18].
• Oracle Cloud at Customer: This is another one popular hybrid cloud ser-
vice provider. It provides mostly on-demand service, in its own cloud system.
Unlike Azure, AWS, and GCP, this provider does not allow its software to
execute in virtual instances for any operation. But it runs on metal servers;
Oracle also offers this kind of service. Oracle cloud is also very easy to run its
apps on-premise on the cloud [18].
• IBM: IBM merged all of its cloud services, called IBM Cloud. It possesses
more than 170 types of services for public cloud and on-premise. These ser-
vices are not only limited to bare metal hosting and virtualized mode, con-
tainers, and server less computing, DevOps, AI/ML, HPC, and blockchain.
It also offers to do lift and shift on-premise apps, executing on IBM plat-
forms [18].
• Cisco Cloud Center: Cisco is popular for private cloud that also offers hybrid
solutions via its partner. Cisco Cloud Center is more secured to manage and
deploy the applications in different data centers in both private and public
cloud environments. Cisco’s partner networks are Google, CDW, Accenture,
and AT&T. Google is the biggest partner among them. It offers the hybrid
connectivity and their solutions [18].
• VMware vCloud Suite: VMware provides vendor for virtualized services.
It is relatively new than that of other service providers. VMware has the
vSphere hypervisor. Customers can run in some known public clouds or
their own data centers or cloud provider partners. These cloud providers
are able to run vSphere on-premise that creates a stable hybrid cloud infra-
structure [19].
1.3.1 Cloud Storage and Backup Benefits

Protection of the confidential data is very difficult. Automatic backup of cloud storage is
flexible. It also provides data security.
Microsoft Azure is very effective in SaaS. Whereas, Google Cloud is strong in AI [18].
Table 1.1 gives a comparison among them.
1.3.2 Pros and Cons of Different Service Providers

All the cloud service providers have their own pros and cons. Their make themselves a suit-
able choice for different purposes. Here, the advantages and disadvantages are described for
all the providers. Table 1.2 provides a comparative study on this.
Table 1.1 Comparison between AWS Outpost, Microsoft Azure Stack, and Google Cloud Anthos.
AWS Outpost Microsoft Azure Stack Google Cloud Anthos
Amazon has a huge tool set The customer can run in Google has come to the
and that too is rapidly their own data center. cloud market later. So,
growing. No service Azure tries to incorporate it does not have that
providers can match with that. It provides the much level of focus to
with it. But the pricing facility of hybrid cloud incorporate the customers.
is bit puzzling. Though [19]. But the strength is its
providing service for A customer can replicate technical efficiency. Some
hybrid or public cloud is his environment in Azure of its efficient tools are
not amazon’s primary focus Stack. This is very useful in applicable in data analytics,
thus incorporation of cloud case of backup disaster and machine learning, and
services with on-premise for cutting cost. deep learning.
data is not in top priority
[20]. They primarily focus
on public cloud.
Table 1.2 Pros and cons between AWS Outpost, Microsoft Azure Stack, and Google Cloud Anthos.
Vendor Strength Weakness
AWS Outpost 1. Dominant market position 1. Managing cost
2. Extensive, mature offerings 2. Very difficult for using
3. Effective use in large organizations 3. Options are overwhelming
Microsoft 1. Second largest service provider 1. Poor documentation
Azure Stack 2. Coupling with Microsoft software 2. Management tooling is incomplete
3. Set of features is vast
4. Provides Hybrid cloud
5. Open source supported
Google Cloud 1. Designed to serve for cloud-native 1. Enters late in IaaS market
Anthos enterprises 2. Less services and features
2. Provides portability and allows 3. Not focused for enterprise
open source
3. Huge discounts and suitable
contracts
4. Expertise in DevOps
1.3.2.1 AWS Outpost

The strongest strength of Amazon is its effectiveness in public cloud. They provide services
through the world for its public cloud infrastructure. This cloud provider is very popular
because of its varieties operational scope. AWS provides different kind of services. It also
has a large network for worldwide data centers. The “Gartner” reported that this provider
is the most mature and enterprise-ready. It also has capabilities to govern a large amount of
resources and customers. But the weakness is its cost. Customers face difficulty to under-
stand its cost structure. It is also difficult to manage the costs while running a large volume
of workloads.
1.3.2.2 Microsoft Azure Stack

Microsoft provides on-premises software—SQL Server, Windows Server, SharePoint,
Office, .Net, Dynamics Active Directory, etc. The reason of its success is most of the
enterprises uses Windows and its related software. As Azure is tightly coupled with its
other software applications, the enterprises, that use many Microsoft software, they
find Azure as a suitable platform. This is how it builds good relationship with their
existing customers. They also provide a remarkable discount on variety of services to
their existing customer. But, Gartner also reported some faults in their some of the
platforms [21].
1.3.2.3 Google Cloud Anthos

AWS and Azure offer the Kubernetes standard which is developed by Google. GCP is expert
in machine learning and Big Data analytics. It provides huge offers on that. It also provides
offers in load balancing and considerable scale. Google is also efficient knowledge about
different data centers and quick response time. Google stands in third in the field of market
share [21]. But, it is rapidly increasing its offers. As per Gartner, clients choose GCP as a
secondary provider than that of primary provider.
1.3.3 Review on Storage of the Providers

1.3.3.1 AWS Outpost Storage
• SSS to EFS: The storage services of AWS include its Elastic Block Storage
(EBS), Simple Storage Service (S3), and Elastic File System (EFS) for persistent
block storage, object storage, and file storage, respectively. It also provides
some new innovative products for storage that includes the Snowball and
Storage Gateway. Snowball is a physical hardware device, whereas Storage
Gateway creates a hybrid storage environment.
• Database and archiving: Aurora is a compatible database of SQL by
Amazon. It consists of different services like DynamoDB NoSQL database,
Relational Database Service (RDS), Redshift data warehouse, ElastiCache
in-memory data store, Neptune graph database, and Database Migration
Service. Amazon also offers long term storage known as Glacier. It is having
very low charges [20].
• Storage services: The storage services of Microsoft Azure include Queue
Storage, Blob Storage, File and Disk Storage for large-volume workloads, and
REST-based object storage of unstructured data respectively. Data Lake Store
is another storage that is used for big data applications.
• Extensive database: This extensive database provides three SQL-based
options. They are Database for MySQL, SQL Database, and Database for
PostgreSQL. Data Warehouse service is also provided as well. The services
are Table Storage for NoSQL and Cosmos DB. Its in-memory service is Redis
Cache and the hybrid storage service is Server Stretch Database. Those are
designed for the organizations that use Microsoft SQL Servers [22]. Unlike
AWS, Microsoft offers an actual Site Recovery service, Archive Storage, and
Backup service.
1.3.3.2 Google Cloud Anthos Storage

• Unified storage and more: GCP has enormous level of storage services. The
unified object storage service is cloud storage. It also provides persistent disk
storage. It also offers a Transfer Appliance which is a similar kind of AWS
Snowball and online transfer services.
• SQL and NoSQL: GCP possesses the SQL-based Cloud and also provides
a relational database known as Cloud Spanner. Cloud Spanner is designed
for critical and complex workloads. It also provides NoSQL. They are Cloud
Datastore and Cloud Bigtable. No backup services and archive services are
provided.
Table 1.3 Comparison between VMware Microsoft Amazon AWS.

Category VMware Microsoft Amazon AWS
Delivery mode Very simple Easy to follow Very easy
Ability to apply the Cost-effective Estimated cost Very affordable, $32 to
technology virtualization was around $255 per month [19]
solution, manage $4.99 per
to virtualize the month [19]
X86 computer
architecture
Integration It is an Edge PC Computes engine Web application,
with other Virtualization, for networking, website and database
applications Workstation 12.5 virtual storefront.
Pro, Fusion 8.5 - machines,
Windows on Mac®, SQL databases,
Workstation 12 storage,
Player- streamlined containers,
PC Virtualization security, API
for Business integration, etc.
Security Secure virtual box is Reliable Tight
possible to create,
manages files, using
SSL, SSH, etc.
Operating system Many operating Windows 8 and Both Linux and
and mobile systems like Windows 10 Windows.
compatibility Windows, Linux Able to compute, storage,
and Mac, etc. database, networking,
and content delivery.
Upgrades On demand Products Able to run
available at less updates.
price.
Service-level Azure Cloud Easy
agreements provides
Container
services
speedily and in
simple way.
Training/support Auditing, monitoring/
logging, storage creating
Scalability and Vendor is dependable
vendor reliability and revenue growth is
stable for Elastic Cloud
Compute (EC2) and
database usage [19]
Another random document with
no related content on Scribd:
Yet to go on might—probably would—spell utter disaster to my
peace of mind, and make shipwreck of my honor.
Hour after hour passed and I seemed to draw no nearer to a
conclusion. But at length the glimmerings of a solution of the
problem began to draw in my mind. If I could but find Stanley Audley
I could cut myself adrift from the mystery and try to forget Thelma as
speedily as possible. This I determined honestly to try to do, and I
think I felt better and happier for the resolution. What I failed to
realize was the strength of the feelings that had me in their grip. And
ever and anon, like an inducement of hope, came the resolution of
Thelma’s declaration that Stanley could never return to her. In that
case—but I resolutely tried to push away from me the thoughts that
crowded into my mind.
Next day, after spending a couple of hours at Bedford Row with my
partner, Hensman, I set out on my first inquiry regarding Stanley
Audley.
I took a taxi to the house in Half Moon Street in which he had lived,
and there saw Mr. Belton, the proprietor.
He was a tall, bald-headed man in grey trousers and morning coat
and nothing could disguise the fact that he was a retired butler. “Yes,
sir,” he said in reply to my inquiry, “Mr. Stanley Audley lived here for
nearly two years. But he went abroad a short time ago, as I wired to
you, sir.”
“Well, the fact is, Mr. Belton, he’s disappeared,” I said.
“Disappeared!” echoed the ex-butler.
“Yes, I wonder if I may glance at his rooms.”
“Certainly, sir. But they are let again. Colonel Mayhew is out, so we
can go up. Mr. Audley sent all his things to store when he left, but I
was away at the time, so I don’t know where they went to.” He took
me to a well-furnished front sitting-room on the first floor.
“Do you recollect that he had a lady visitor—a tall, handsome, dark-
eyed young lady, whose name was Shaylor?”
“Certainly, sir. A young lady came once or twice to tea, but I don’t
know her name. And—well to tell you the truth, sir, his movements
were often very curious.”
“How?” I asked, with sudden interest.
“Well, he would walk out without any luggage sometimes, and then a
week later I would hear from him telling me to send on his letters to
some Poste Restante abroad. Once it was in Paris, another time at
Geneva and twice in Madrid. It always struck me as very curious that
he traveled without any luggage—or if he had any, he never brought
it here.”
“Curious,” I said. “Then he was a bit of a mystery?”
“He was, sir. That’s his photograph there, on the mantleshelf,” and
he pointed to a photograph in a small oval ebony frame.
To my amazement it was the picture of a man I had never seen in my
life.
“But that round-faced man isn’t Stanley Audley!” I exclaimed.
“Excuse me, sir, but it is,” was the ex-butler’s polite assertion. “He
lived here nearly two years.”
“He is not the Stanley Audley for whom I am searching, at any rate,”
I said.
“Well, he is the only Mr. Audley that my wife and I have had here.”
Suddenly I recollected that in my wallet I had a snap-shot of Thelma
on her skis which I had taken up on the Allmendhubel. I drew it out
and showed it to him.
“Ah! sir, that’s not the young lady who visited Mr. Audley. That’s a
young lady who came twice, or perhaps three times to see Mr.
Graydon.”
“What is Mr. Graydon like?” I asked eagerly.
In reply he gave me a very accurate description of Thelma’s
husband.
“Who, and what is Mr. Graydon?” I asked. “Tell me, Mr. Belton, for
much depends upon the result of this inquiry.”
“He’s a young gentleman very well connected—nephew of a certain
earl, I believe. He had the rooms above for about nine months, and
was very friendly with Mr. Audley.”
“And did he make mysterious journeys?”
“Yes, sometimes—but not very often.”
“Had he any profession?” I inquired.
“No. I understand that his father, who was a landowner in Cheshire,
left him with a very comfortable income. My wife and I liked him, for
he was a quiet, rather studious young fellow, though often at Mr.
Audley’s invitation he went out of an evening and did not return till
the early hours. But now-a-days with those dance clubs going, most
young men do that.”
“Well, Mr. Belton, may I see Mr. Graydon’s room?” I asked. In
response, he took me up to the next floor, where the sitting-room and
bedroom were even cosier and better furnished than the rooms
below.
“Mr. Graydon, when he left, laughingly said that he might be married
soon, but if he didn’t marry he’d come back to us. He told my wife
that he was going on a yachting trip to Norway with some friends,
and afterwards he had to go to Montreal to visit some relatives.”
“But the curious fact is that the man I knew as Audley is none other
than the man you know as Graydon!” I said.
“That’s certainly very mysterious, sir. Mr. Graydon must have
assumed Mr. Audley’s name,” Belton said.
“The whole affair is a complete mystery,” I remarked. “I wish you’d
tell me more that you know concerning this Mr. Graydon. What was
his Christian name, by the way? And when did you last see him?”
“Philip. He left us last September.”
“And the young lady who came to see him?”
“Oh! She was certainly a lady. Indeed, I rather fancied that I had
seen her several years ago, and that with her mother she once came
as guest of old Lady Wentbrook, in whose service I was. But I was
not quite sure, and I could not, of course, inquire. At any rate, she
was a lady, of that there could be no mistake.”
“And Mr. Graydon was a gentleman?”
“Certainly, sir. But I can’t vouch for Mr. Audley. They were friends—
and that’s all I know.”
“You had certain suspicions about Audley, and were not sorry when
he gave up his rooms?”
“Yes, sir, you’re quite right, I was.”
“And how about Graydon?”
“We were very sorry when he left, sir. My wife liked him immensely.
But she always said that he was somehow under the influence of Mr.
Audley.”
“Did you ever meet a Mr. Harold Ruthen?” I asked.
And from my wallet I took another snap-shot which showed him with
a party of skaters on the rink.
The ex-butler scrutinized it closely and replied:
“Yes. He’s been here. He was a friend of Mr. Audley’s. But I don’t
think that was his name. I believe he was called Rutley, or some
such name?”
“Did Mr. Graydon know him?”
“No, sir. Not to my knowledge. He came here once and stayed with
Mr. Audley while Mr. Graydon was up in Scotland shooting. But we’ll
go down below and show the photograph to my wife. She has a
better memory than I have.”
So we went into the basement, where I had a long conversation with
Mrs. Belton, a typical retired servant of the better class, shrewd and
observant.
That conversation definitely established several amazing facts which
served to make the mystery of Stanley Audley deeper and more
sinister than ever. It was clear—
(1) That Philip Graydon had, for some reason we could not
fathom, taken the name of Stanley Audley, while Audley
had passed as Graydon.
(2) That the movements of the two men were uncertain and
mysterious.
(3) That Harold Ruthen, also known as Rutley, was associated
with both Stanley Audley and the man Philip Graydon.
(4) That Thelma had married the man who, passing as Philip
Graydon, was really Stanley Audley!
After that amazing revelation I passed along Half Moon Street, in the
winter darkness, to Piccadilly in a state of utter bewilderment.
CHAPTER VI
THE HAM-BONE CLUB
A few days later a client of ours named Powell for whom we were
conducting a piece of rather intricate business concerning a
mortgage of some land in Essex, invited me to join himself and his
wife at dinner at the Savoy.
Our table was in a corner near the orchestra and the big restaurant
was crowded. Sovrani, the famous maître d’hôtel knew all three of us
well and we dined excellently under his tactful supervision. After
dinner Mrs. Powell, a pretty young woman, exquisitely gowned,
suggested a dance in the room below. We went there and danced
until about half-past ten when Powell said:
“Let’s go to the Ham-bone.”
“The Ham-bone,” I echoed. “What on earth is that?”
“Oh!” laughed Mrs. Powell, “it is one of London’s merriest Bohemian
dance clubs. The male members are all artists, sculptors or literary
men, and the female members are all girls who earn their own living
—mannequins, secretaries, artists’ models and girl journalists. It is
screamingly amusing. Quite Bohemian and yet high select, isn’t it,
Harry?”
“I’ve never heard of it,” I said.
“Well, one gets a really splendid dinner there for half-a-crown,
though, of course, you get paper serviettes, and for supper after the
hours, you men can have a kipper—a brand that is extra special—
and a drink with it,” she went on.
“Yes, Leila,” laughed her husband. “The place is unique. Half the
people in ‘smart’ society, men as well as women, want to become
members, but the Committee, who are all well-known artists, don’t
want the man-about-town: they only want the real hard-working
Bohemians who go there at night for relaxation. Burlac, the sculptor,
put me up.”
The novelty of the idea attracted me, so we went in a taxicab to an
uninviting looking mews off Great Windmill Street, behind the Café
Monico in Piccadilly Circus. Walking up it, we passed through a
narrow swing-door, over which hung a dim feeble light and a big
ham-bone!
Up a precipitous flight of narrow stone steps we went until we
reached a little door where a stout ex-sergeant of police smiled
recognition upon my host, placed a book before him to sign and
relieved us of our coats.
In a room above a piano was being played by someone who was
evidently an artist and dancing was in progress.
The place might have been a cabaret in the Montmartre in Paris. I
thought I knew London’s night clubs fairly well—the Embassy, Ciro’s,
the Grafton, the Mayfair, the Royalty, the Twenty, Murray’s, Tate’s,
the Trippers, the Dainty, and others—but when I entered the big
whitewashed dancing room I found myself looking on a scene that
was a complete novelty to me.
The room was long and narrow. The walls were painted in stripes
representing oaken beams and set around them were many small
tables. The floor was filled with merry dancers, among whom I
recognized many people well-known in artistic and social circles.
Some of the men wore dinner jackets and many of the women were
in beautiful evening dress, but smart clothes evidently were regarded
as a non-essential, for a large proportion of the men wore ordinary
lounge suits.
As we stood watching the scene a tall, elderly man rose from a table
and cried:
“Hulloa! Leila! What a stranger you are!”
My hostess smiled and waved recognition, whereupon her friend—a
portrait painter whose reputation was world-wide, bowed over her
hand and said:
“Well, only fancy! It is really delightful that you should return to us!
We thought we’d lost you after you married!”
“My dear Charlie,” she laughed—for it was a rule in the Ham-bone
that every member addressed every one else by his or her Christian
name, and “Charlie” was a Royal Academician—“I am an old
Hamyardian: I was one of the first lady members.”
“Of course. You’ll find Marigold here. I’ve just been chatting with her.
She’s round the corner, over yonder. But she’s funny. What’s the
matter with her? Do you know?” he added in a low, serious voice.
“No, I didn’t know there was anything wrong,” replied my hostess.
It was easy to realize that here in this stable converted into a club
was an atmosphere and an environment without its like in London or
elsewhere. The denizens of that little circle of Bohemia cared for
absolutely nothing and nobody outside its own careless world whose
boundaries were Chelsea and the Savoy Club.
Ordinary social distinctions were utterly and completely ignored.
Gayety was supreme and in the merry throng I caught sight within a
few minutes of a well-known London magistrate before whom I had
often pleaded as a Solicitor, a famous scientist, the millionaire owner
of a great daily paper. Several leading members of the Chancery
Bar, an under-secretary of State and quite a sprinkling of young
scions of patrician families.
They were men and women of the intellectual type who cared
nothing for the vicious joys of the ordinary night club. They came in
frank enjoyment of dancing and music and the fried kippers, as
custom decreed, in order to comply with the kill-joy law that ordained
that they must eat if they wanted a drink! Everything, apparently, was
free and easy gaiety. Yet it was at least as difficult to become a
member of the Ham-bone as to gain admission to any of the most
exclusive clubs along Pall Mall. Money was no sort of passport: only
personality, ability or the true inborn spirit of Bohemianism could
open the portals of the Ham-bone.
The “master of ceremonies” was a well-known landscape painter,
whom every one addressed as “George,” a smart figure in the brown
velvet jacket of his profession. He chaffed and joked with every one
in French, revealing a side of his nature certainly unsuspected by the
general public to whom he usually presented a grave and austere
front. But this was the key-note of the Ham-bone: every one seemed
to “let himself go” and the stilted social etiquette of our ordinary world
seemed as far off as if we had been in Limehouse or Poplar.
I was dancing with Mrs. Powell, when, suddenly, she halted before a
small table in a corner where there sat alone a beautiful dark-haired
girl in a smartly cut dance-frock of black charmeuse.
“Mr. Yelverton,” she said, “will you let me introduce you to my dearest
friend, Marigold Day?” And to the girl she said, “Marigold, this is Mr.
Rex Yelverton, the gentleman of whom I recently spoke to you.”
Somberly dressed, her white neck and bare arms in vivid contrast
with her dead-black frock, she was almost wickedly beautiful. Her
well-dressed hair, across which she wore a bandeau of golden
leaves, was dark; her scarlet mouth was like the curling underleaves
of a rose, her lips with the true arc-de-cupidon so seldom seen, were
slightly apart, and between them showed strong white teeth. Her
eyes were large and deeply violet and they held a fascination such
as I had seldom before seen.
“We’ll be back presently,” said Mrs. Powell, as we slipped again into
the dance. “I want to have a chat with you.”
“Who’s that?” I asked, as soon as we were a few feet away.
“Oh, that’s Marigold. We are fellow-members here. She was in
business with me before I married. Isn’t she very good-looking, don’t
you think?”
“Beautiful,” I declared.
“Ah, I see,” laughed my partner. “You are like all the other men. They
all admire her, and want to dance with her. But Marigold is a queer
girl: I can never make her out in these days. Once she was very
bright and merry, and always gadding about somewhere with a man
named Audley. Now there’s a kink somewhere. She accepts no
invitations, keeps herself to herself, and only on rare occasions
comes here just to look on. A great change has come over her. Why,
I can’t make out. We were the closest of friends before I married, so
I’ve asked her the reason of it all, but she will tell me absolutely
nothing.”
“Audley,” I gasped. “Where is she at business?”
“At Carille’s, the dressmakers in Dover Street. She’s a mannequin,
and I was a typist there,” she replied. “And now Mr. Yelverton, you
know what was my business before I married,” she added, with a
laugh.
“Pretty boring, I should say, showing off dresses to a pack of
unappreciative old cats,” was my remark.
“Boring isn’t the word for it,” Mrs. Powell declared, “I couldn’t have
stood her work. You should see our clients—uneducated, fat, coarse,
war-rich old hags who look Marigold up and down, and fancy they
will appear as smart as she does in one of Monsieur Carille’s latest
creations. How Marigold sticks at it so long I can’t make out. She
ought to be awarded the prize medal for patience. I could never
amble about over that horrid grey carpet and place my neck, my
elbows and hands at absurd angles for the benefit of those ugly old
tabbies—no matter what salary I was paid!”
At that moment we found ourselves before the table where her
husband was seated, smoking and drinking coffee with Sava, the
young Serbian who was perhaps the greatest modern caricaturist.
Belgravia is good; Bohemia is better; the combination of both is
surely Paradise! Sava’s conversation was as perfect as his
caricatures: he had seen life in every capital in Europe and was a
born raconteur. For a time he held us engrossed with his witty
comments on the men and matters of half-a-dozen countries, all of
which he knew to perfection.
Never have I seen so truly fraternal a circle as that little backwater of
Bohemianism. Every one was at his ease: there was no such a thing
as being a stranger there. The fact that you were there—that some
member had introduced you and vouched for you—broke down all
barriers and men who had never before met and might never meet
again met and chatted as freely as if they were old friends and with
an utter disregard of all the vexing problems of wealth, rank,
profession and precedence.
Presently my hostess took me back to the mannequin in black whom
I new realized must be wearing a copy of one of the famous man-
dressmaker’s latest creations.
“Mr. Yelverton wants a partner, Marigold,” my companion exclaimed
gayly, whereupon her friend smiled and rising at once, joined me in a
fox-trot with an expression of pleasure upon her face. She was a
splendid dancer.
“Mrs. Powell has told me of your acquaintance with Mr. Audley,” I
said, after a few minutes of the usual ball-room chat. “I wonder if it is
the same man I know. He used to live in Half Moon Street.”
She clearly resented the question. “Why do you ask?” she
demanded.
“Because I’ve lost sight of my friend of late,” I replied.
“Well, Mr. Audley did live in Half Moon Street, but he has gone
away,” she replied. And I thought I detected a hint of tragedy upon
her face.
CHAPTER VII
IN THE WEB
As we danced Marigold told me something more about herself. She

lived, I found, with three other business girls at a boarding house in
Bayswater, going by tube to Dover Street each day. She had met
Audley and for a time they had been rather friendly, seeing a good
deal of each other. I guessed, though of course she did not tell me,
that the friendship bade fair to ripen into something deeper. Then
Audley had suddenly disappeared.
As our dance ended Mrs. Powell came up and we all went up the
narrow wooden staircase to the balcony where, as we enjoyed our
Bohemian supper, we could watch the dancing below.
It was just before midnight, when the fun was fast and furious and
the “Hamyardians,” as the merry circle call themselves, were
enjoying themselves in the wildest and most nonsensical fashion,
that Marigold Day, glancing at her wrist watch, declared that she
must go. I went down with her to the door.
“Can’t you tell me some more about Audley?” I asked just before she
entered her taxi.
She shook her head. “Don’t ask me, please,” she said and she
entered the taxi and was driven away towards Bayswater.
“Well, what do you think of Marigold?” asked Mrs. Powell, as I
resumed my seat at the supper table.
“She’s altogether charming, of course,” I replied, “but rather—well, I
don’t quite know the word. I should almost say mysterious: at any
rate she seems to be troubled about something and trying to hide it.”
“That’s it, exactly,” declared my hostess. “During the past few months
she seems to have become an entirely different girl. As you know,
we were the closest of friends. She seems to live in constant dread
of something, but she absolutely refuses to tell me what it is. Indeed,
she declares there is nothing wrong, but that is nonsense. No one
who knew her six months ago could fail to realize that something is
very wrong indeed.”
“Do you know anything about her friend, Mr. Audley,” I ventured to
ask.
“Not very much,” said Mrs. Powell. “Of course, I have met him.
Marigold was getting very fond of him, I believe, but she will not talk
about him.”
Powell came up and declared it was time to go and I had no
opportunity of questioning Mrs. Powell any further, much as I wished
to do so. However, I determined to see her again and also to meet
Marigold Day and see whether either of them could give me further
details about Audley. Was he the real Audley? I wondered, or the
man who had taken his name.
A few days later I received a letter from Mrs. Shaylor inviting me to
go to Bexhill.
I was in two minds about accepting. I wanted to see Thelma—
wanted to help her and certainly did not want to lose touch with her
as I might if I refused to go. But was it wise?
Of course, inclination conquered prudence and I went. I found that
she and her mother lived in a pretty red-roofed, red-brick detached
house, with high gables, and a small garden in front. It stood in
Bedford Avenue, close to the Sackville Hotel and facing the sea.
Mrs. Shaylor, a pleasant, grey-haired woman of a very refined type,
greeted me warmly and thanked me cordially for what I had done for
her daughter in Mürren, while Thelma expressed her delight at
seeing me again.
I got a chance during the morning of speaking to Mrs. Shaylor alone
and asked her if Thelma had heard anything more of her husband.
“Not a word,” was Mrs. Shaylor’s reply. “It is a most disastrous affair
for her, poor girl. The suspense and anxiety are killing her.”
“She does not look so well,” I replied. I had, in fact, been struck by
the change in the girl. She was paler and thinner and it was evident
the strain was telling on her rather heavily.
“I understand you did not know very much of Mr. Audley,” I said.
“Very little indeed, unfortunately,” was Mrs. Shaylor’s reply. “Thelma
met him when she was staying with her aunt at the Majestic at
Harrogate, and they became friendly. He appeared to have
considerable means for he gave Thelma some very beautiful jewelry.
He came down here once, saw me, and asked if he might marry her.
He told me certain things about his relations in India, and she
seemed so entirely devoted to him that I gave my consent to their
marriage in three months. But, judge my surprise when a fortnight
later they were married secretly and left next day for Switzerland for
their honeymoon.”
“Then you really know very little of him, Mrs. Shaylor?” I asked.
“Very little indeed. It was a most foolish and ill-advised marriage. He
seems to have lied to her here and then deserted her.”
“I must say I liked what I saw of him,” I said, “and I wonder whether
we are right in thinking that he really deserted her in the ordinary
meaning of the word. It looks like it, of course, but it has occurred to
me, though I have only very slight grounds to go on, that he is being
kept away from her by some influence at which we cannot guess. He
really seemed devoted to her and genuinely sorry to have to leave
her.”
“Well, she certainly seems devoted to him and will not hear a word
against him. But what can one think under the circumstances?”
The drawing-room opened on to a wide verandah and across the
promenade we could see the rolling Channel surf beating upon the
beach. The winter’s day was dull and boisterous and now and again
sheets of flying spray swept across the promenade.
“He pretended to me that he was an electrical engineer,” I remarked,
“but I have found out that the firm for whom he said he worked
knows nothing of him.”
“That is what he also told me. But I have reason to believe that he is
in fact a young man of considerable fortune. Yet, if so, why has he
deserted poor Thelma?”
“I am doing my level best to find him, Mrs. Shaylor,” I said. “Some
very great mystery enshrouds this affair, and I have, in your
daughter’s interest, set myself to solve it.”
“I’m sure all this is extremely good of you,” she said, gratefully. “We
are only women, and both of us powerless.”
I paused for a moment. Then I said:
“I really came down here, Mrs. Shaylor, to put several direct
questions to you. I wonder if you will answer them and thus lighten
my task. I am a solicitor, as perhaps you already know.”
“Certainly. What are they?”
“Has your daughter ever known a man named Harold Ruthen?”
The lady’s face changed, and her brows contracted slightly. “Why do
you ask that?” she asked.
“Because it has a direct bearing upon the present situation.”
“Well—yes. I believe she has, or had, a friend of that name. A man
who lives in Paris.”
“Was he a friend of Audley’s?”
“Not to my knowledge.”
“Have you ever heard of a girl named Marigold Day—a mannequin
at Carille’s?”
“Never.”
I paused. Then I bent towards her and said, very earnestly, “Has it
ever struck you, Mrs. Shaylor, that your daughter knows just a little
more concerning Stanley Audley than she has yet told us?”
“Why do you ask that question?” she inquired.
“Well—because somehow it has struck me so,” I said. “And I will go
a little further. I believe she knows where her husband is, but—for
some reason or other—fears to betray him!”
“Is that your suspicion?” she asked, in a low strained voice.
“Yes,” I replied.
“Mr. Yelverton,” she said very slowly. “I admit that it is mine also! I’ve
questioned Thelma time after time, but she will tell me nothing—
absolutely nothing!”
“Are there any more facts you can tell me—anything to throw further
light upon these strange circumstances?” I asked her.
“No,” was her reply. “I’m afraid I know nothing else. Thelma is
worried. I feel terrified lest the real truth—whatever it may be—
concerning her husband, be disclosed.”
Thelma came in and we talked of other matters. She made great fun
of my position as her “temporary husband” at Mürren and seemed in
better spirits than when I came down.
After luncheon we went for a stroll together through the driving
health-giving breeze to Cooden Beach, and then back for tea.
Thelma wore a serviceable golf suit, thick brogues and carried a
stick, while her Airedale “Jock” ran at our side.
On the way I told her of my adventure at the Ham-bone Club. She
was much interested in the queer pranks of the Hamyardians and to
find out how much she knew, I told her about Marigold Day: in fact I
deliberately “enthused” about her. I watched her closely, but it was
evident Marigold’s name meant nothing to her. Then I went on the
more open tack and tried to get some further facts from her. It was in
vain: she seemed as determined to keep her knowledge to herself as
I was to get at the truth.
At last, as we neared the house, I made a direct attack.
“Now look here, Thelma,” I said, “do be frank. You know where
Stanley is, don’t you?”
She went pale: it was evident that it had never struck her that I might
guess at the truth.
“Why do you say that?” she asked sharply.
“Because I am certain Stanley has enemies and wants help.”
“Enemies!” she said, with an attempt to laugh “why should he have
enemies? What do you mean?”
“All that I have said. Cannot you trust me? If your husband is in
hiding for some unknown reason I should not betray him.”
“I have promised to say nothing,” she said blankly. “I cannot break
my promise.”
“Why does he not return to you?”
“There is a reason—he never can. We must live apart in future.”
“Why?”
She shrugged her shoulders, and after a few moments of hesitation
replied—
“There are certain facts, Mr. Yelverton, that I am forbidden by Stanley
to disclose. I have told you that we cannot be united again. That is
all. Please make no further inquiries.”
“But I will. You have been left in my care,” I asserted.
“If you do!—if you do it—it may be at your peril,” she declared, in a
hard unnatural voice, looking curiously at me as she opened the
gate. “Recollect, Mr. Yelverton, that my words are a warning.”
“But why?” I cried.
“I—I unfortunately cannot tell you,” was her reply, and we re-entered
her charming home together.
I returned to London more mystified than ever. The dual personality
of Stanley Audley, combined with the fact that his wife undoubtedly
knew of his whereabouts; her steadfast determination not to disclose
one single fact, and the strange threats I had heard Ruthen utter, all
combined to puzzle me beyond measure.
For a couple of days I did my best to attend to business, but
constantly I found my mind dwelling on the mystery of Stanley
Audley. I could not concentrate on legal problems and most of my
work fell on Hensman’s shoulders.
On the third night, after my visit to Bexhill, when I returned to my
rooms from the office, I found, lying upon my table, a typewritten
note which had been delivered that afternoon. It bore the
Hammersmith postmark.
Tearing it open I read some lines of rather indifferent typing, as
follows:—
“You have formed a friendship with Mrs. Thelma Audley. I
warn you that such friendship, if continued, will be at the
cost of your own life. Divert your love-making into another
direction. I have no personal animosity against you but
you are placing yourself in the way of powerful interests,
and you will be removed if necessary.”
I read and re-read this strange message. Thelma’s warning leaped
to my mind. Was there, then, a real risk to myself in the strange coil?
Then something—sheer obstinacy I suppose—came to my help and
I declared to myself that I would go ahead with my self-imposed task;
that nothing—least of all mere cowardice—should induce me to give
it up.
CHAPTER VIII
DOCTOR FENG’S VIEW
I am not going to deny that at first that strange warning perturbed me

a good deal. After all, I make no claim to be a hero and not even a
hero likes threats of death, even though they be anonymous. At the
same time, I never proposed, even in thought, to give up my quest.
For, whether I wished it or not, I could not shake myself free of
Thelma’s influence: my day-dreams were themselves on the fancy
that some day, in some way, she would be free. More and more I
began to think that she had married Audley so suddenly under an
overwhelming girlish impulse; perhaps her mind had been made up
by some story he had told her to justify haste and secrecy. If this
were really so, would her love survive desertion and a separation
which she herself apparently regarded as permanent? It would be
strange, indeed, if it did.
So, through the dark March days that followed, I worked at the office
half the day, while the remainder I devoted to seeking traces of the
mysterious young man who had lived in Half Moon Street under the
name of Graydon.
Mrs. Powell and her husband had been suddenly called abroad. But
Marigold Day was an obvious source of possible information and to
make further inquiry of her I wrote asking her to dine with me one
evening at the Cecil.
She accepted, and we ate our dinner at one of the tables set in the
window of the big grill-room overlooking the Embankment. She again
wore her plain black dress which enhanced the whiteness of her
arms and shoulders and laughed merrily at me across the table as
we chatted over dinner.
I hesitated to refer to Audley directly after the conversation of our
previous meeting, but I asked her suddenly whether she happened
to know a man named Harold Ruthen.
“Harold Ruthen?” she echoed, “Yes, but why do you ask?”
“Because he was a friend of Audley’s,” was my reply. “Do you
happen to know him?”
“Certainly. I saw him only a few days ago. He’s looking for Audley—
he believes he is in Paris.”
“Now, I wonder if the Mr. Audley you know is the same man as my
friend. Will you describe him?”
She did so, and the description made it clear that he was indeed
Thelma’s husband.
“Yes,” I said. “He is no doubt the same.”
“He was well-known at the Ham-bone, where every one called him
Stanley,” she said. “But I can’t think why he disappeared and has
never written to me. A girl told me that he’d married. But I don’t
believe it.”
“Why not?”
“For the simple reason that he had asked me to marry him,” was the
startling reply.
“Was Ruthen on very friendly terms with him?”
“Yes. But Stanley did not like him. He used to tell me that Ruthen
was not straight, and I know he avoided him whenever he could. I
suppose we all hate most those we fear most.”
“Why do you say that?” I asked in some surprise at her philosophy.
“Well,” she said, “I always had a suspicion that Stanley went in fear
of Ruthen. Why, I don’t know.”
“That’s curious. What made you think so?”
“From certain remarks he once let drop.”
“Then Audley may be hiding purposely from that fellow?” I
exclaimed, as I recollected that queer conversation between Ruthen
and Thelma.

Machine Learning Techniques and Analytics For Cloud Security Advances in Learning Analytics For Intelligent Cloud Iot Systems 1St Edition Chakraborty

Uploaded by

Copyright:

Available Formats

Machine Learning Techniques and Analytics For Cloud Security Advances in Learning Analytics For Intelligent Cloud Iot Systems 1St Edition Chakraborty

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Machine Learning Techniques and Analytics For Cloud Security Advances in Learning Analytics For Intelligent Cloud Iot Systems 1St Edition Chakraborty

Uploaded by

Copyright:

Available Formats

Full download test bank at ebookmeta.

Machine Learning Techniques and Analytics for

Download More ebooks from https://ebookmeta.com

Machine Learning Approach for Cloud Data Analytics in

Machine Learning for Intelligent Multimedia Analytics

Machine Learning and IoT for Intelligent Systems and

Deep Learning Approaches to Cloud Security Deep

Machine Learning for Cloud Management 1st Edition Kumar

Machine Learning for Intelligent Multimedia Analytics

Advances in Machine Learning Deep Learning based

Architecting Data and Machine Learning Platforms:

Advances in Learning Analytics for Intelligent Cloud-IoT Systems

Series Editor: Dr. Souvik Pal and Dr. Dac-Nhuong Le

Submission to the series:

Wiley Global Headquarters

Limit of Liability/Disclaimer of Warranty

Library of Congress Cataloging-in-Publication Data

Cover images: Pixabay.Com

Printed in the USA

2 Recognition of Differentially Expressed Glycan Structure

Part II: Cloud Security Systems Using Machine Learning

6.3.4.3 Characteristics of IDS 109

8.4 Machine Learning for Cloud Computing 156

9.9 Experimental Setup 182

Part III: Cloud Security Analysis Using Machine Learning

11.1 Introduction 215

11.2 Cloud Computing Overview 216

12.4 Machine Learning for Cloud Security 245

13.3.4 Textual-Based Attacks (NLP) 276

Part IV: Case Studies Focused on Cloud Security 313

15.5.4 Preemptible VMs 332

16.5.2.2 Blob Storage 347

17.4.2.2 AHV 367

Part V: Policy Aspects 379

18.4.5 Measuring Policy Effectiveness in the User Interaction Graph 403

Part I: Conceptual Aspects of Cloud and Applications of Machine Learning

use of cloud in education, finance, etc., is described, the chapter concludes

Part II: Cloud Security Systems Using Machine Learning Techniques

Part III: Cloud Security Analysis Using Machine Learning Techniques

also presented. Then, various studies on adversarial attacks on cloud-based

Part IV: Case Studies Focused on Cloud Security

Part V: Policy Aspects

RCC Institute of Information Technology, Kolkata, West Bengal, India

Keywords: PaaS, SaaS, IaaS, SLA, agility, encryption, middleware, AI

*Corresponding author: moudeb@gmail.com

• Scalability: IT services are not restricted to offline resources anymore,

• Security: Cloud service providers use many security mechanisms like

1.2 Hybrid Cloud

• Workload distribution by portability.

1.2.2 Why Hybrid Cloud is Required?

HYBRID CLOUD MODEL

Figure 1.1 General architecture of hybrid cloud.

• Application programmer always requires support for edge technologies.

1.2.3 Business and Hybrid Cloud

Hybrid Cloud Market - Growth Rate by Region (2020 - 2025)

Regional Growth Rates

Figure 1.2 Market trend of hybrid cloud [14].

1.2.4 Things to Remember When Deploying Hybrid Cloud

1.3 Comparison Among Different Hybrid Cloud Providers