Heterogeneity, High Performance Computing, Self-Organization and the Cloud.

• Superior document: Palgrave Studies in Digital Business and Enabling Technologies Series
• Author: Lynn, Theo.
• TeilnehmendeR: Morrison, John P.; Kenny, David.
• Place / Publishing House: Cham : : Springer International Publishing AG,, 2018.; ©2018.
• Year of Publication: 2018
• Edition: 1st ed.
• Language: English
• Series: Palgrave Studies in Digital Business and Enabling Technologies Series
• Physical Description: 1 online resource (183 pages)

Table of Contents

• Intro
• Preface
• Acknowledgements
• Contents
• Notes on Contributors
• List of Abbreviations
• List of Figures
• List of Tables
• Chapter 1: Addressing the Complexity of HPC in the Cloud: Emergence, Self-Organisation, Self-Management, and the Separation of Concerns
• 1.1 Introduction
• 1.2 Cloud Computing
• 1.3 High Performance Computing
• 1.4 HPC and the Cloud
• 1.5 Heterogeneous Computing
• 1.6 Addressing Complexity in the Cloud through Self-* Design Principles
• 1.7 Application Scenarios
• 1.7.1 Oil and Gas Exploration
• 1.7.2 Ray Tracing
• 1.7.3 Genomics
• 1.8 Conclusion
• 1.9 Chapter 1 Related CloudLightning Readings
• References
• Chapter 2: Cloud Architectures and Management Approaches
• 2.1 Introduction
• 2.2 Cloud Architecture
• 2.2.1 Infrastructure Organisation
• 2.2.1.1 The Switch-Centric Model
• 2.2.1.2 The Server-Centric Model
• 2.2.2 The Cloud Management Layer
• 2.2.2.1 OpenStack
• 2.2.2.2 Google Kubernetes
• 2.2.3 The Service Delivery Layer
• 2.3 Transitioning to Heterogeneous Clouds
• 2.3.1 Resource Management
• 2.3.2 Resource Abstraction
• 2.4 The CloudLightning Approach
• 2.4.1 Infrastructure Organisation
• 2.4.2 Hardware Organisation
• 2.4.2.1 Resource Abstraction
• 2.4.3 The Cloud Management Layer
• 2.4.3.1 CL-Resource Discovery
• 2.4.3.2 The CL-Resource Selection
• 2.4.3.3 Resource Acquisition
• 2.4.3.4 Coalition Lifecycle Management
• 2.4.3.5 Self-Organisation Agent
• 2.4.3.6 Classification of vRack Managers
• 2.4.3.7 vRack Manager Activities
• 2.4.4 Service Delivery Model
• 2.4.5 Advanced Architecture Support
• 2.4.5.1 Auto-Scaling
• 2.4.5.2 High Availability
• 2.4.5.3 Data Locality
• 2.4.5.4 Dynamic VPN Creation for Blueprint Service Execution
• 2.5 Conclusion
• 2.6 Chapter 2 Related CloudLightning Readings
• References.
• Chapter 3: Self-Organising, Self-Managing Frameworks and Strategies
• 3.1 Introduction
• 3.2 Key Concepts
• 3.3 Augmenting the CloudLightning Architecture
• 3.4 Self-Organisation and Self-Management in CloudLightning Architecture
• 3.4.1 Directed Evolution
• 3.4.1.1 The Goal State
• 3.4.1.2 Cell State
• 3.4.1.3 pRouter State and pSwitch State
• 3.4.1.4 vRM State
• 3.4.1.5 Steering by the Cell
• 3.4.1.6 Steering by the pRouter
• 3.4.1.7 Steering by the pSwitch
• 3.4.2 Self-Management Mechanisms
• 3.4.2.1 Mechanism to Send Metrics from a vRM to pSwitch
• 3.4.2.2 Mechanism to Send Metrics from a pSwitch to pRouter
• 3.4.2.3 Mechanism to Send Metrics from pRouter to Cell
• 3.4.2.4 Mechanism to Send Weights from Cell to pRouters
• 3.4.2.5 Mechanism to Send Weights from pRouters to pSwitches
• 3.4.2.6 Mechanism to Send Weights from pSwitch to vRMs
• 3.4.2.7 A Mechanism in the Cell to Modify Local Behaviour in an Effort to Respond to Impetus Provided by the Directed Evolution and Metrics Coming from Attached pRouters
• 3.4.2.8 A Mechanism in a pRouter to Modify Local Behaviour in an Effort to Respond to Impetus Transmitted by the Cell and Metrics Coming from Attached pSwitches
• 3.4.2.9 A Mechanism in a pSwitch to Modify Local Behaviour in an Effort to Respond to Impetus Transmitted by its pRouter and Metrics Coming from Attached vRMs
• 3.4.2.10 A Mechanism in a vRM to Modify Local Behaviour in an Effort to Respond to Impetus Transmitted by its pSwitch and Metrics Coming from its vRack
• 3.4.2.11 Sample Events that Trigger the Transmission of Metrics at each Level in the Hierarchy
• 3.4.2.12 Sample Events that Trigger the Transmission of Weights at Each Level in the Hierarchy
• 3.4.3 Self-Organisation Mechanisms
• 3.5 CloudLightning SOSM Strategies
• 3.5.1 Self-Management Strategies.
• 3.5.1.1 An Example Self-Management Scenario
• 3.5.2 Self-Organisation Strategies
• 3.5.2.1 An Example Self-Organisation Scenario
• 3.6 Conclusion
• 3.7 Chapter 3 Related CloudLightning Readings
• Chapter 4: Application Blueprints and Service Description
• 4.1 Introduction
• 4.2 Representative Application Lifecycle and Resource Management Frameworks
• 4.3 CloudLightning Stakeholders and Associated Concerns
• 4.4 The CloudLightning Approach Based on Separation of Concerns
• 4.4.1 CloudLightning Requirements
• 4.4.2 Separation of Concerns
• 4.4.2.1 Application Lifecycle Management
• 4.4.2.2 Resource Lifecycle Management
• 4.5 The CloudLightning Gateway Architecture
• 4.5.1 Gateway Service Architecture
• 4.5.2 Service Decomposition
• 4.5.3 Interaction with the SOSM System
• 4.5.3.1 Resource Discovery
• 4.5.3.2 Resource Release
• 4.6 The CloudLightning Blueprint Extensions
• 4.6.1 CloudLightning Brooklyn Extensions
• 4.6.2 CloudLightning Abstract Blueprint
• 4.6.3 CloudLightning Blueprint
• 4.7 Example of Application Creation and Deployment
• 4.8 Conclusion
• 4.9 Chapter 4 Related CloudLightning Readings
• References
• Chapter 5: Simulating Heterogeneous Clouds at Scale
• 5.1 Introduction
• 5.2 Cloud Simulation Frameworks
• 5.3 CloudLightning Simulator
• 5.3.1 Architecture and Basic Characteristics of the Parallel CloudLightning Simulation Framework
• 5.3.2 SOSM Engine
• 5.3.2.1 Power Consumption Modelling
• CPU Power Models
• Combined CPU-Accelerator Power Models
• 5.3.2.2 Memory, Storage, and Network Modelling
• 5.3.2.3 Application Models
• 5.3.2.4 Execution Models
• 5.4 Experimental Results
• 5.5 Conclusion
• 5.6 Chapter 5 Related CloudLightning Readings
• References
• Chapter 6: Concluding Remarks
• References
• Index.