Intro
Preface
Organization
Contents
Cloud and High-Performance Computing
File Access Patterns of Distributed Deep Learning Applications
1 Introduction
2 Related Work
3 Characterizing the I/O Patterns Models of DDL Applications
3.1 Software Stack DL
3.2 File Access Pattern
4 Experimental Data-extraction for File Access Pattern Modelling Characterization
4.1 Experimental Environment
4.2 Mechanisms Used to Characterize File Access Patterns
4.3 Characterization of File Access Patterns to the CIFAR-10 Dataset

4.4 Characterization of File Access Patterns to the MNIST Dataset
5 Conclusions
References
A Survey on Billing Models for Cloud-Native Applications
1 Introduction
2 Systematic Literature Review
3 Main Findings and Discussion
4 Conclusions and Research Opportunities
References
Performance Analysis of AES on CPU-GPU Heterogeneous Systems
1 Introduction
2 Background
2.1 AES Algorithm
2.2 Characterization of Heterogeneous Systems
2.3 Related Work
3 Previous Implementations of AES
3.1 AES for Multicore CPU
3.2 AES for Single-GPU and Multi-GPU

4 AES for CPU-GPU Heterogeneous Systems
5 Experimental Results
6 Conclusions and Future Work
References
Network Traffic Monitor for IDS in IoT
1 Introduction
2 Network Traffic Monitor Architecture
3 Deployment and Testing
3.1 Creating Topology Elements. OpenFlow Switch
3.2 Creating Links Between Components
3.3 Connecting the Monitor
3.4 Creating Host 1 and Host 2
3.5 Connecting Host 1 and Host 2
4 Creating SDN Controller and Traffic Sniffer
5 Conclusions and Future Work
References
Crane: A Local Deployment Tool for Containerized Applications

1 Introduction
2 Container Management Architecture Precedents
2.1 SWITCH
2.2 COCOS
2.3 Lightweight Kubernetes Distributions
3 Design Evolution of Crane
3.1 Instances Load Balancing
3.2 Container Automatic Scaling
3.3 Detected Implementation Problems
4 Conclusions and Future Work
References
Machine and Deep Learning
Multi-class E-mail Classification with a Semi-Supervised Approach Based on Automatic Feature Selection and Information Retrieval
1 Introduction
2 Background
3 Research Methodology
3.1 Description of the Dataset

3.2 Labeling of Documents
3.3 Email Indexing
3.4 Feature Selection Strategies
3.5 Retrieval of E-mails
3.6 Generation of the Classification Models
4 Experiments
5 Conclusions
References
Applying Game-Learning Environments to Power Capping Scenarios via Reinforcement Learning
1 Introduction
2 The RLlib and Gym Frameworks
2.1 RLlib
2.2 Gym
3 RL for Resource Management
4 Casting a Power Capping Scenario with Gym
4.1 Defining States
4.2 Defining Actions and Rewards
5 Experimental Results
5.1 Analysis Under Different Power Caps