VM Migration via Passive Counterpart Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for migrating virtual machines (VMs) between data centers are inadequate as they do not provide granularity in VM migration, are not seamless, and do not ensure continuous data protection, especially during site failures or link disruptions.
Innovation Solution
The system involves freezing an active VM at one data center, creating a synchronized passive counterpart, and then restarting the VM at another data center, using snapshot technology to efficiently migrate and synchronize VMs while maintaining continuous data integrity and failover capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all VMs on a volume are frozen and the entire volume is copied to the second site, then data consistency and completeness are ensured, but migration granularity is lost and downtime increases
Solution Approach 1:
The patent divides the volume into individual VM-level units for migration. Instead of freezing and copying the entire volume, the system identifies and migrates specific VMs individually, allowing other VMs to continue running. This segmentation enables granular control over migration scope and minimizes overall downtime.
Solution Approach 2:
The patent creates a passive counterpart of the target VM at the second site before actually migrating the active VM. This preliminary creation of the destination environment allows the system to prepare the migration path in advance, ensuring data consistency is established beforehand while the actual VM migration can proceed quickly without freezing operations.
2Quantity of substance
If the entire volume is copied to ensure all VMs are migrated, then data completeness is ensured, but migration flexibility and selectivity are reduced
Solution Approach 1:
The system segments the migration process at the VM level within the volume. Administrators can selectively migrate individual VMs or groups of VMs based on business needs, rather than being forced to migrate the entire volume. This provides flexibility to migrate only critical or specific VMs while leaving others operational.
Solution Approach 2:
The patent enables extraction of specific VMs from the larger volume context for independent migration. The system identifies and extracts individual VMs or selected groups from the volume, allowing them to be migrated separately without affecting other VMs in the same volume. This extraction capability provides precise control over migration scope.
3Reliability
If VMs are frozen for migration to ensure data consistency, then data protection is improved, but operational continuity is disrupted
Solution Approach 1:
The system performs preliminary synchronization of the passive VM counterpart with the active VM before actual migration. This preliminary action establishes data consistency and protection in advance, allowing the actual VM migration to proceed without freezing operations and maintaining operational continuity.
Solution Approach 2:
The patent maintains continuous operation of VMs during the migration process. By creating passive counterparts in advance and migrating individual VMs without freezing the entire system, the useful action of VM operation continues uninterrupted. The system ensures data protection through continuous synchronization rather than through freezing.
4Reliability
If synchronous volume replication is used to ensure fast data transfer, then data consistency is improved, but migration time and system complexity increase
Solution Approach 1:
The system segments the data transfer process at the VM level rather than performing monolithic synchronous volume replication. Each VM can be synchronized and migrated independently, allowing parallel processing and reducing overall migration time while maintaining data consistency through targeted synchronization operations.
Solution Approach 2:
The patent creates passive VM counterparts at the destination site before migrating active VMs. This preliminary creation allows the system to establish the migration path and synchronize data in advance, enabling faster subsequent migration operations without requiring lengthy synchronous replication processes.
Data Source
AI summary
Migrating an active VM from a first data center to a second data center having a passive counterpart of the active VM includes freezing the active VM at the first data center, creating an active VM at the second data center that corresponds to the passive counterpart, and restarting the active VM at the second data center. Migrating an active VM from a first data center to a second data center may also include waiting for the passive counterpart to be synchronized with the active VM at the first data center prior to converting the passive counterpart. Creating the active VM at the second data center may include providing a snapshot of a volume containing the passive VM.