Virtual Machine Replication via Datastore Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current disaster recovery systems face challenges in efficiently replicating and recovering virtual machines across geographically separate datacenters, particularly in integrating external replicators that operate at the virtual machine level within a framework defined for datastore replication.
Innovation Solution
The system pairs and maps datastores between a source and target server, designating them as storage-only to manage replication through a disaster recovery orchestrator, which schedules and replicates virtual machines by generating snapshots and renaming configuration files, enabling VM-based replication while adapting external replicators to work within the datastore-defined interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If datastore replication framework is used for virtual machine replication, then integration with existing disaster recovery systems is improved, but replication efficiency and flexibility are worsened
Solution Approach 1:
The system segments the replication process by introducing a virtualization layer that separates VM management from datastore replication. The external replicator operates at the VM level while the internal replicator handles datastore-level operations, allowing independent optimization of each layer and improving overall replication efficiency without sacrificing framework integration.
Solution Approach 2:
The patent introduces an intermediary virtualization layer between the external replicator and the datastore replication framework. This intermediary manages the translation between VM-level replication requests and datastore-level operations, enabling efficient VM replication while maintaining compatibility with existing disaster recovery systems.
2Reliability
If entire datastore replication is performed, then data protection is improved, but recovery time and resource consumption are worsened
Solution Approach 1:
The system extracts and replicates only the specific virtual machines that need to be recovered, rather than replicating entire datastores. This selective extraction approach reduces the amount of data that must be transferred and recovered, significantly reducing recovery time while maintaining adequate data protection for the selected VMs.
Solution Approach 2:
The patent implements partial replication by allowing users to select specific VMs for replication instead of requiring full datastore replication. This partial action approach reduces the scope of replication operations, decreasing both the time required for recovery and the computational resources consumed during the process.
3Adaptability or versatility
If external replicators operate at VM level, then replication flexibility is improved, but compatibility with datastore-defined interface is worsened
Solution Approach 1:
The patent adds a new dimension to the replication architecture by introducing a virtualization layer that operates above the datastore level. This allows external replicators to work at the VM level with full flexibility while the underlying datastore replication framework maintains its existing interface definitions, effectively resolving the compatibility issue through dimensional separation.
Data Source
AI summary
A first server in, for example, a primary datacenter, may be paired with a second server in, for example, a backup datacenter. The first server may include a first datastore, which includes a plurality of virtual machines, and the second server may include a second datastore. The first and second datastores may then be mapped to each other. The plurality of virtual machines included in the first datastore are then, without any user prompting, replicated at the second datastore. Therefore, when the first server becomes unavailable, the virtual machines included in the first datastore may still be accessible at the second server.