Virtual Storage Dataset Recovery Using Tiered Consistency Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage systems face inefficiencies in data management and reliability due to unnecessary write operations and lack of centralized control over storage processes, leading to reduced reliability and increased wear on storage devices.
Innovation Solution
Implementing a direct-mapped flash storage system where higher-level processes, such as the operating system, manage data allocation and storage operations directly, eliminating the need for lower-level storage controllers to perform redundant processes, and utilizing non-volatile RAM for quick data buffering and power failure protection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If storage controllers perform data management operations, then storage operations can be automated, but device complexity increases and reliability decreases due to redundant write operations
Solution Approach 1:
The patent extracts the storage controller from the data path by implementing a direct-mapped flash storage system where the operating system directly manages flash storage devices without intermediary storage controllers. This eliminates the storage controller's redundant write operations and associated reliability issues while maintaining automation at the operating system level.
Solution Approach 2:
The patent introduces a translation layer that maps logical block addresses from the operating system directly to physical flash storage locations, eliminating the need for storage controllers as intermediaries. This translation mechanism enables direct OS-to-flash communication while maintaining address translation functionality.
2Ease of operation
If storage controllers manage data allocation, then data management is centralized, but write operations become redundant and device wear increases
Solution Approach 1:
The patent removes storage controllers from the data management hierarchy, allowing the operating system to directly allocate and manage flash storage resources. This extraction eliminates the redundant write operations that storage controllers perform when copying data between buffers and flash memory, thereby reducing device wear.
Solution Approach 2:
The operating system performs self-service data management by directly accessing and managing flash storage devices without external storage controller intervention. The OS handles allocation, addressing, and data placement decisions, eliminating the need for storage controllers to perform duplicate management functions.
3Productivity
If traditional storage controllers are used, then storage operations can be performed, but data integrity is compromised during power failures
Solution Approach 1:
The patent implements preliminary action by using non-volatile RAM to buffer data before it reaches flash storage, and by maintaining a clean shutdown flag that records the last known good state. Upon power restoration, the system can quickly determine whether to commit buffered data or rollback to the last stable state, preventing corruption from incomplete write operations.
Solution Approach 2:
The patent provides beforehand cushioning through non-volatile RAM that preserves buffered data during power failures, and through a shutdown flag mechanism that marks the state of flash storage at the last clean shutdown. This cushioning protects against data loss by maintaining a known good state that the system can restore if power is lost during write operations.
Data Source
AI summary
Data recovery in a virtual storage system, including: detecting, within storage provided by a first tier of storage of the virtual storage system, data loss within a dataset, wherein recovery data for the dataset is stored in a second tier of storage; determining a recovery point for the dataset up to which a consistent version of the dataset is recoverable from the recovery data stored in the second tier of storage; and restoring, within the storage provided by the first tier of storage of the virtual storage system, the consistent version of the dataset.