Cluster Data Migration via Script-Orchestrated State Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for backing up applications on cloud computing systems are limited, as snapshots can only be used to restore applications on clusters with the same configuration, preventing migration to different node setups or versions.
Innovation Solution
A computer-implemented method that orchestrates the execution of scripts to back up and restore data across virtual computing instances, allowing data to be migrated to clusters with different configurations, including those on different cloud systems or versions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If snapshots are used to back up applications on cloud computing systems, then the backup process is simple and fast, but the snapshots can only be used to restore applications on clusters with the same configuration
Solution Approach 1:
The patent segments the backup process into two distinct phases: (1) capturing data state information from the source cluster, and (2) restoring that information to a target cluster. This segmentation allows the backup to be configuration-agnostic, enabling restoration on clusters with different configurations while maintaining simple snapshot-based capture mechanics.
Solution Approach 2:
The patent introduces an intermediary layer of data state information that acts as a bridge between source and target clusters. Instead of directly copying snapshots which require configuration matching, the system captures application state, configuration, and data information, then uses this intermediary representation to restore applications on different cluster configurations.
2Ease of operation
If virtual disk snapshots are copied to restore applications, then the restoration process is straightforward, but the new cluster must have the same number of nodes, pools, and affinities as the previous cluster
Solution Approach 1:
The patent changes the parameters being backed up from fixed virtual disk configurations to dynamic data state information including application state, configuration metadata, and data content. This parameter transformation allows restoration on clusters with different node counts, pool structures, and affinity settings while maintaining straightforward restoration operations.
Solution Approach 2:
The backup system is designed to be universal by capturing application state information that can be restored across multiple cluster configurations. The same backup artifact can restore applications on clusters with different numbers of nodes, different pool assignments, and different scheduling affinities, making the restoration process universally applicable.
3Reliability
If cluster configuration must match for snapshot restoration, then data integrity is maintained, but migration between different cloud systems or versions is prevented
Solution Approach 1:
The patent performs preliminary capture of complete application state information including configuration metadata, application state, and data content before migration. This preliminary action creates a self-contained backup package that preserves data integrity while being independent of the source cluster configuration, enabling subsequent migration to different cloud systems or versions.
Solution Approach 2:
Instead of copying virtual disks which are tightly coupled to specific cluster configurations, the patent copies application state information in a configuration-agnostic format. This copying approach maintains data integrity through faithful state reproduction while enabling migration across different cluster environments by decoupling the backup from source configuration constraints.
Data Source
AI summary
One embodiment of a method for migrating data includes performing one or more first operations to orchestrate execution of one or more first scripts in one or more first virtual computing instances, wherein the execution of the one or more first scripts causes data associated with the one or more first virtual computing instances to be backed up to at least one first disk, copying the data from the at least one first disk to at least one second disk, and performing one or more second operations to orchestrate execution of one or more second scripts in one or more second virtual computing instances, wherein the execution of the one or more second scripts causes the data to be restored from the at least one second disk to the one or more second virtual computing instances.