Instant Storage Volume Cloning via Redundant References
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data center virtualization technologies face challenges in efficiently replicating storage volumes for creating exact duplicates, which is time-consuming and hinders rapid testing, development, and fault tolerance.
Innovation Solution
A system with redundant storage volumes synchronizes all writes across both volumes, allowing instantaneous creation of clones with continuous data replication, enabling quick access to replicated storage without the need for new data creation, and maintaining these replicas in diverse locations for fault tolerance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional storage volume replication methods are used, then data copies can be created with high accuracy, but the replication process takes excessive time
Solution Approach 1:
The system pre-creates redundant storage volumes and maintains them in a suspended or standby state before they are needed. When a clone is requested, the pre-positioned redundant volume can be immediately activated and associated with the computing instance, eliminating the need for time-consuming data copying at the moment of need.
Solution Approach 2:
Instead of copying actual data blocks, the system creates redundant volume references or pointers that point to the original storage volume. These references are immediately available and can be activated instantly, providing the effect of instant cloning without the overhead of physical data duplication.
2Reliability
If storage volumes are replicated for fault tolerance, then system reliability is improved, but the complexity of managing multiple copies increases
Solution Approach 1:
The system introduces a management layer or intermediary component that handles the complexity of redundant volume management. This intermediary manages the creation, suspension, activation, and association of redundant volumes with computing instances, shielding users from the underlying complexity while ensuring high availability.
Solution Approach 2:
The system extracts the redundancy management functionality into a separate, dedicated component or service. This allows the core storage and computing operations to remain simple while the specialized redundancy management system handles fault tolerance independently, reducing the complexity burden on the main system.
3Reliability
If redundant storage volumes are maintained in diverse locations, then fault tolerance against regional failures is improved, but the cost of maintaining multiple distributed copies increases
Solution Approach 1:
The system merges multiple redundant volume references into a unified management structure. Instead of maintaining completely separate physical copies in diverse locations, the system uses reference pooling where multiple redundant volumes can point to the same underlying data storage, reducing actual data duplication while maintaining geographic distribution for fault tolerance.
Data Source
AI summary
Techniques are described for allocating computing storage capacity to customers of a provider network. Storage capacity that is allocated to a customer is backed up with a replica of the allocated storage capacity. A request is received for a copy of contents of the allocated storage capacity. Responsive to the indication, the requested copy is provided and is configured to reference the replica when the requested copy is accessed.