Cloud Controller Service Instance Discovery and Binding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud controllers face challenges in managing existing service environments and service instances created or modified by systems other than themselves, as they are unaware of these instances and their configurations, leading to difficulties in performing lifecycle management and ensuring compliance with service level agreements.
Innovation Solution
The implementation of a 'brown-to-manage' process that includes a discovery phase to gather information about service instances, a binding phase to infer and associate a service model with policies and actions, and a management phase to manage the lifecycle of these instances, enabling the cloud controller to manage previously unknown or modified service instances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If cloud controllers only manage service instances they created themselves, then management control and awareness are maintained, but the ability to manage existing service environments and instances created by other systems is lost
Solution Approach 1:
The discovery phase is performed before management operations to proactively gather information about service instances. This preliminary action ensures that the cloud controller has necessary configuration data before attempting to manage instances, resolving the contradiction by preparing information in advance rather than reacting to information gaps during management operations.
Solution Approach 2:
The patent introduces an intermediary discovery and binding mechanism that bridges the gap between service instances created by other systems and the cloud controller. This intermediary process translates and standardizes information from various sources into a format the cloud controller can manage, enabling versatility while maintaining information integrity through structured data transformation.
2Productivity
If cloud controllers perform discovery and binding operations on existing service instances, then lifecycle management capability is improved, but system complexity and operational overhead increase
Solution Approach 1:
The management process is segmented into distinct phases: discovery phase for information gathering, binding phase for model association, and management phase for lifecycle operations. This segmentation reduces complexity by breaking down the overall task into manageable, standardized steps that can be executed systematically rather than as a monolithic complex process.
Solution Approach 2:
The system performs self-service through automated discovery and binding operations. The cloud controller automatically discovers service instances, infers their models, and binds them to appropriate service templates without requiring manual intervention. This automation improves productivity while the standardized self-service nature of the operations keeps complexity manageable through repetition and standardization.
3Reliability
If cloud controllers infer service models from existing instances, then compliance with service level agreements is ensured, but the time and resources required for model determination increase
Solution Approach 1:
The system creates copies of service models from existing service instances through the binding phase. Instead of repeatedly analyzing instances to determine their models, the system infers the model once and creates a reusable service model copy that can be applied to multiple operations. This copying approach ensures compliance through consistent model application while reducing time loss by avoiding repeated inference processes.
Solution Approach 2:
The binding phase provides feedback mechanisms that validate inferred service models against actual service instance behavior. This feedback loop ensures compliance with service level agreements by verifying that the inferred model accurately represents the service instance, while the learned information from feedback reduces future inference time through pattern recognition and model refinement.
Data Source
AI summary
In some examples, information of a service instance provisioned or modified by a first system distinct from a management system is received, where the provisioning or modification of the service instance by the first system is initially unknown to the management system. As part of a binding phase, a model of the service instance is inferred. The management system manages a lifecycle of the service instance using the model and the policy.