Workload Migration Controller for External Topology Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud controllers struggle to manage and migrate workloads that were provisioned by systems other than themselves, as they lack awareness of the service models and topology of these workloads, leading to challenges in lifecycle management and migration.
Innovation Solution
A workload migration controller is employed to discover and manage workloads with a brown-to-manage process, inferring the topology model and performing migrations while maintaining component relationships, using a two-step migration process of coarse and delta movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a cloud controller manages only workloads it provisioned itself, then management control and awareness of service models are maintained, but the controller cannot manage workloads provisioned by other systems
Solution Approach 1:
The system performs preliminary discovery actions by sending discovery packets to workload components before migration. This preliminary action gathers topology information and service model data in advance, allowing the cloud controller to understand workloads it did not originally provision, thereby resolving the contradiction between managing external workloads and maintaining information awareness.
Solution Approach 2:
The system introduces an intermediary discovery mechanism that acts as a mediator between the cloud controller and external workloads. This intermediary process enables the controller to obtain topology and service model information from workloads provisioned by other systems without requiring direct knowledge or integration with those provisioning systems.
2Speed
If workload migration is performed in a single step, then migration speed is improved, but the complexity of ensuring service continuity and handling failures increases
Solution Approach 1:
The migration process is segmented into multiple phases: initial migration of workload components, establishment of connectivity, and subsequent failover. This segmentation allows each phase to be managed independently with appropriate error handling and rollback mechanisms, reducing overall process complexity while maintaining migration speed through parallel operations within each segment.
Solution Approach 2:
The system performs preliminary actions by pre-configuring target infrastructure and establishing connectivity before actual workload migration. This preliminary setup reduces the complexity of the main migration event and enables faster execution by having the target environment ready to receive workload components immediately.
3Loss of time
If all workload components are migrated simultaneously, then migration time is reduced, but the risk of service disruption and failure impact increases
Solution Approach 1:
Workload components are segmented into migration batches that can be transferred in parallel. This segmentation allows the system to maintain service continuity by keeping non-migrated components operational while migrating others, reducing overall service disruption risk while minimizing total migration time through concurrent operations.
Solution Approach 2:
The system establishes cushioning measures by creating redundant connectivity paths and maintaining backup configurations before migration begins. This beforehand cushioning ensures that if migration failures occur, services can continue operating on existing configurations while corrective actions are taken, thereby protecting service continuity during parallel migration operations.
Data Source
AI summary
In some examples, a system migrates, responsive to a request, a workload comprising components and relationships among the components as represented by a topology model, the migrating comprising migrating the workload from the source infrastructure to a target infrastructure, and migrating components of the workload from the source infrastructure to the target infrastructure.