Data Storage Controller Parallel Version Erasure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face inefficiencies in managing multiple versions of data over time, leading to suboptimal space and performance utilization.
Innovation Solution
A data storage device with a controller that receives identification of different data versions from a host, stores them in erasable memory areas, and manages their parallel erasure, reading, and movement to cold data areas, optimizing storage and access efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple versions of data are stored in different data storage devices, then data version management is achieved, but storage space utilization deteriorates and device complexity increases
Solution Approach 1:
The patent combines multiple versions of data into a single data storage device by organizing them in a version tree structure within the same address space. Different versions are stored at different logical block addresses (LBAs) in the same device, allowing version management without requiring multiple separate storage devices, thereby improving storage space utilization while maintaining version management capability
Solution Approach 2:
The patent segments the storage space into different regions for hot data and cold data, and further segments versions within each region. By dividing the address space into version-specific segments and organizing them in a tree structure, the system efficiently manages multiple versions without wasting storage space, resolving the contradiction between version management and storage utilization
2Quantity of substance
If multiple versions of data are stored in the same data storage device, then storage space utilization improves, but data access speed deteriorates due to increased complexity in locating and retrieving specific versions
Solution Approach 1:
The patent performs preliminary organization of data versions by building a version tree structure during data writing operations. The controller pre-calculates and pre-positions version data in optimized locations within the address space, and maintains a version map that directly maps version identifiers to physical locations. This preliminary structuring enables fast access to specific versions without requiring complex searches during retrieval operations
Solution Approach 2:
The patent introduces a version map and address translation layer as an intermediary between the host and the physical storage locations. This intermediary structure translates logical version identifiers into physical block addresses, enabling the host to access specific versions quickly without directly navigating the complex internal storage structure. The version map acts as a lookup table that resolves version access requests in constant time
3Reliability
If invalid versions of data are spread across a large number of metablocks, then data integrity is maintained, but storage efficiency deteriorates and erase operations become less efficient
Solution Approach 1:
The patent merges invalid version data into consolidated regions within the address space rather than scattering them across many metablocks. By combining invalid versions into specific cold data regions or designated invalid data zones, the system maintains data integrity through proper version tracking while enabling efficient erasure operations. The controller can erase consolidated invalid data regions in larger units, improving erase efficiency while preserving valid version data through the version tree structure
4Productivity
If parallel erasure of multiple data versions is implemented, then productivity improves, but device complexity increases due to coordination requirements
Solution Approach 1:
The patent segments the erasure operation into independent parallel tasks by organizing invalid versions as separate entries in the version tree structure. Each invalid version can be identified and erased as an independent unit, allowing the controller to distribute erasure operations across multiple processing channels simultaneously. The version tree structure provides clear boundaries and identifiers for each version segment, enabling parallel processing without complex inter-dependency coordination
Data Source
AI summary
A data storage device and method for host-assisted efficient handling of multiple versions of data are provided. In one embodiment, a data storage device is provided comprising a memory and a controller. The controller is configured to receive, from a host, identification of different versions of data that are to deleted together, store the different versions of the data in areas of the memory that are erasable in parallel; receive, from the host, a command to erase the different versions of the data; and erase the different versions of the data in parallel. Other embodiments are provided.