Edge Computing and Capability-Oriented Architecture

Cover
Half Title
Title Page
Copyright Page
Contents
Foreword
SECTION I: Edge computing software fundamentals
	1. Edge computing fundamentals
		1.1. Finding the edge
			1.1.1. Edge computing models
			1.1.2. Defining edge computing
		1.2. Edge computing characteristics
			1.2.1. Compute in context
			1.2.2. Capability oriented
			1.2.3. Centralized management, distributed compute
			1.2.4. Secured
			1.2.5. Heterogeneity
		1.3. Edge computing scenarios
			1.3.1. IoT scenarios
			1.3.2. Hardware as a service
			1.3.3. Hybrid scenarios
			1.3.4. 5G scenarios
			1.3.5. Edge computing in a pandemic
		1.4. Summary
	2. Edge computing factors
		2.1. Edge computing networking
			2.1.1. Software-defined network (SDN)
			2.1.2. Sensor networks
			2.1.3. 5G
			2.1.4. Long-distance networks
		2.2. Edge devices
			2.2.1. Edge host
			2.2.2. Virtualization
			2.2.3. Software stack
			2.2.4. Device security
			2.2.5. Intelligent devices
		2.3. Fleet management
			2.3.1. Clustering
			2.3.2. Identities and access control
			2.3.3. Configurations and policies
		2.4. Perspectives and uniformity
			2.4.1. Human players
			2.4.2. Multiple perspectives
			2.4.3. Uniformity
		2.5. Summary
SECTION II: Edge computing design patterns
	3. Edge to cloud
		3.1. Data collection patterns
			3.1.1. Data collection through messaging
			3.1.2. Data collection through ingress endpoints
			3.1.3. Bulk data transportation and in-context data processing
			3.1.4. Data pipeline on cloud
		3.2. Remoting patterns
			3.2.1. Connected consumer devices
			3.2.2. Online collaborations
			3.2.3. Cloud-based relays
			3.2.4. Multi-party computations
		3.3. Compute offloading patterns
			3.3.1. Multi-level offloading
			3.3.2. Bursting to cloud
			3.3.3. Adaptive offloading
			3.3.4. Dynamic delegation
		3.4. Summary
	4. Cloud to edge
		4.1. Edge acceleration patterns
			4.1.1. Smart CDNs
			4.1.2. EdgeWorkers
		4.2. Cloudlet pattern
			4.2.1. AWS Wavelength
			4.2.2. Azure Edge Zones
		4.3. Edge functions pattern
			4.3.1. AWS IoT Greengrass
			4.3.2. Azure IoT Edge
			4.3.3. Apache OpenWhisk
			4.3.4. Function host management at scale
			4.3.5. Edge PaaS paradox
		4.4. Cloud compute stack on edge
			4.4.1. AWS Outputs
			4.4.2. Azure Stack
			4.4.3. OpenStack
		4.5. Edge native compute stack
			4.5.1. Uplifting on-premises datacenters
			4.5.2. Template-based compute stack
			4.5.3. Multi-access edge computing framework
		4.6. Autonomous systems
			4.6.1. From automatic to autonomous
			4.6.2. The human factor
			4.6.3. Autonomous system architecture
			4.6.4. Cloud-assisted robotics
		4.7. Summary
	5. Kubernetes on edge
		5.1. An anatomy of a Kubernetes cluster
			5.1.1. Kubelet
			5.1.2. CRI
			5.1.3. OCI
			5.1.4. Device plugin
			5.1.5. CNI
			5.1.6. Cluster API
		5.2. Lightweight Kubernetes clusters
			5.2.1. Minikube
			5.2.2. K3s
			5.2.3. MicroK8s
		5.3. Separating the control plane and the compute plane
			5.3.1. KubeEdge
			5.3.2. OpenYurt
		5.4. Custom kubelet implementations
			5.4.1. Virtual kubelet
			5.4.2. Krustlet
		5.5. Lightweight container runtimes
			5.5.1. CRI-O
			5.5.2. Kata containers
			5.5.3. Windows containers
			5.5.4. Nabla containers
			5.5.5. k3 container solution
			5.5.6. Other noticeable container runtimes
		5.6. Cluster federation
			5.6.1. Federation topologies
			5.6.2. Key challenges
		5.7. Securing Kubernetes clusters on edge
			5.7.1. Container isolation
			5.7.2. Data protection
			5.7.3. Access control
			5.7.4. Security monitoring
		5.8. Summary
	6. Edge native design
		6.1. Edge native applications
			6.1.1. Autonomous bootstrapping
			6.1.2. Adaptive to environmental changes
			6.1.3. Edge high availability
			6.1.4. End-to-end security
			6.1.5. Manageability at scale
		6.2. A model for designing edge native applications
			6.2.1. The OSMP model
			6.2.2. Perspectives
			6.2.3. Views
		6.3. Edge native design
			6.3.1. Bootstrapping
			6.3.2. Adaptive design
			6.3.3. Edge high availability
			6.3.4. End-to-end security practices with the OSMP model
			6.3.5. Manageability
		6.4. Summary
SECTION III: Capability-oriented architecture
	7. Introduction to capability-oriented architecture
		7.1. Capability proxy
			7.1.1. Location transparency
			7.1.2. Hybrid patterns
			7.1.3. Proxy middleware
			7.1.4. Proxy bootstrapper, capability set, capability profile and acquisition agent
			7.1.5. Proxy HA
		7.2. Intention
			7.2.1. Interpreting user intention
			7.2.2. Intention annotations
		7.3. Semantic service discovery
			7.3.1. Semantic service discovery process
			7.3.2. Capability endpoint
			7.3.3. Classic service discovery and description
			7.3.4. Acquisition recipe
			7.3.5. Client-initiated auctions
		7.4. COA compute plane
			7.4.1. Skynet-style computing
			7.4.2. Crowdsourced cloud
			7.4.3. Better computing for everyone
		7.5. Context
			7.5.1. User context
			7.5.2. Activity context
			7.5.3. Device context
		7.6. Summary
	8. COA applications
		8.1. A phone without applications
		8.2. Intelligent applications
		8.3. Zero-touch device provisioning
			8.3.1. Device provisioning process
			8.3.2. Design a generic solution
		8.4. Collaborative computing
		8.5. Context-aware computing
			8.5.1. Device context descriptor
			8.5.2. Adapt to network conditions
			8.5.3. Real-world contexts
		8.6. Approachable AI
			8.6.1. AI inferences
			8.6.2. Custom AI models
		8.7. Browser-based cloud
		8.8. COA – A confession
		8.9. Summary
Index