VM Group Migration Convergence Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for migrating multiple virtual machines (VMs) suffer from split state issues, where some VMs remain on the source platform while others are on the destination, leading to inconsistencies and performance degradation, especially in tiered applications and clustered databases.
Innovation Solution
The solution involves a group migration approach that synchronizes the memory pre-copy convergence of VMs, adjusts dirty rates and network transmit rates, and employs a group-level stun and switch-over protocol to ensure simultaneous convergence and minimize downtime, using techniques like Routed Resume Handshake and Ready to Resume Broadcast to maintain consistency and availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If parallel migration is used to migrate multiple VMs simultaneously, then migration speed and productivity are improved, but split state issues occur causing reliability degradation
Solution Approach 1:
The patent merges the migration processes of multiple VMs into a coordinated group migration operation. By combining individual VM migrations into a unified group migration framework with centralized coordination, the system achieves both parallel execution speed and consistent group state, preventing split states while maintaining high productivity.
2Reliability
If sequential migration is used to ensure group consistency, then reliability is improved, but migration time and loss of time increase
Solution Approach 1:
The patent performs preliminary actions by pre-coordinating migration parameters, establishing convergence time windows, and preparing synchronization mechanisms before actual migration begins. This preliminary coordination enables subsequent parallel migration to proceed efficiently while maintaining group consistency, reducing overall migration time compared to sequential execution.
3Productivity
If memory pre-copy rate is increased to reduce migration time, then productivity is improved, but dirty rate increases causing manufacturing precision degradation
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor dirty rates and memory pre-copy convergence during migration. Based on this feedback, the system dynamically adjusts pre-copy rates and identifies optimal convergence points, ensuring memory consistency is maintained while achieving efficient migration speeds through iterative optimization.
4Manufacturing precision
If dirty rate is reduced to improve memory consistency, then manufacturing precision is improved, but migration time and loss of time increase
Solution Approach 1:
The patent applies dynamic adjustment of pre-copy rates based on real-time monitoring of dirty rates and convergence progress. Rather than using a static low pre-copy rate, the system dynamically optimizes the rate to balance memory consistency requirements with migration time constraints, achieving both goals through adaptive control.
Data Source
AI summary
A plurality of virtual machines (VMs) is migrated from a source group to a destination group in such as way as to achieve consistency and either availability or group preservation. Execution of VMs in the source group is selectively delayed during state migration so that memory transfer of all the VMs in the group will converge roughly at the same time. After VM state transfer to the destination group, execution switch-over is coordinated using different handshake and acknowledgement messages, passed either through a “leader” VM in each group, or directly between source-destination VM pairs.