Flash Storage Snapshot Management via OS-Level Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage systems face inefficiencies in data management and reliability due to redundant write operations and lack of direct control over flash storage processes, leading to increased latency and potential data loss.
Innovation Solution
Implementing a direct-mapped flash storage system where the operating system initiates and controls processes, such as data allocation and garbage collection, without involving lower-level storage controllers, and using non-volatile RAM as a buffer to improve write latency and data persistence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional storage controllers manage flash storage operations, then data management is automated, but write latency increases and unnecessary write operations occur
Solution Approach 1:
The patent extracts the flash storage management functions from the traditional storage controller and places them directly in the operating system. The OS initiates and controls flash storage processes including data allocation, wear leveling, and garbage collection without storage controller intervention, eliminating redundant operations and reducing write latency.
Solution Approach 2:
The flash storage device performs self-management through OS-controlled processes. The system implements self-service mechanisms for data allocation, wear leveling, and garbage collection where the flash storage automatically manages its own operations based on OS directives, reducing the need for complex controller mediation.
2Productivity
If redundant write operations are performed by storage controllers, then data reliability is maintained, but storage efficiency decreases
Solution Approach 1:
The patent removes the redundant storage controller layer that performs unnecessary write operations. By having the OS directly manage flash storage operations, the system eliminates duplicate data handling while maintaining reliability through OS-level control and coordination of write operations.
Solution Approach 2:
The system performs write operations only when necessary, avoiding the excessive redundant writes performed by traditional controllers. The OS selectively initiates write operations based on actual data management needs, performing partial actions only where required rather than redundant full cycles.
3Loss of time
If non-volatile RAM is used as a buffer, then write latency is reduced, but data persistence requirements must be met
Solution Approach 1:
The patent introduces non-volatile RAM as an intermediary buffer between the OS and flash storage. This NVRAM buffer accepts write operations quickly, providing low-latency acknowledgment to the OS, while ensuring data persistence through its non-volatile nature and subsequent flushing to flash storage when appropriate.
Solution Approach 2:
The system uses NVRAM as a cushioning buffer that absorbs write operations before they reach the flash storage. This beforehand cushioning allows the system to acknowledge writes immediately while maintaining data safety in the non-volatile buffer, protecting against potential data loss even before flash write completion.
Data Source
AI summary
An illustrative data storage system captures snapshots of a data structure based on snapshot creation schedules and sets retention periods for the snapshots based on snapshot retention schedules. The data storage system eradicates snapshots based on expirations of the retention periods. In certain examples, the data storage system determines a rule to use to capture a snapshot based on a state of snapshots within one or more lookback periods and/or based on a set of rules each defining a snapshot capture schedule and a snapshot retention schedule.