Snapshot Tree Segmentation for Storage Flexibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage systems have limitations in creating snapshots, including restrictions on the number of snapshots that can be created under a single Logical Disk, requirement for same RAID type, and inability to create mirror clones or snapclones if snapshots already exist, which hinders flexibility and efficiency in data management.
Innovation Solution
The solution allows for the creation of an unlimited number of snapshots under a Logical Disk by splitting existing snapshots into a standalone tree, enabling the creation of mirror clones or snapclones without deleting prior snapshots, and allowing changes in RAID levels without losing existing snapshots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a limited number of snapshots are created under a single Logical Disk, then storage space is saved, but flexibility and operational versatility are reduced
Solution Approach 1:
The patent divides the snapshot management structure into two separate components: a snapshot tree that stores snapshot metadata and a standalone tree that stores actual snapshot data. This segmentation allows the system to maintain an unlimited number of snapshots in the snapshot tree while managing storage space efficiently through the standalone tree, resolving the contradiction between flexibility and storage space utilization.
Solution Approach 2:
The patent introduces a new dimensional structure by separating snapshots into a hierarchical tree format rather than a flat structure. The snapshot tree organizes snapshots in generations, allowing unlimited snapshots to be managed through hierarchical relationships, while the standalone tree provides flat storage for actual data. This dimensional change enables both high versatility and efficient space usage.
2Adaptability or versatility
If all snapshots must have the same RAID type, then data consistency is maintained, but adaptability to different RAID configurations is reduced
Solution Approach 1:
The patent segments the snapshot storage into a snapshot tree for metadata management and a standalone tree for data storage. This allows different RAID types to be used for different snapshots within the same structure, as each snapshot can be independently configured in the standalone tree without affecting others, thus achieving both flexibility and consistency.
Solution Approach 2:
The patent enables parameter changes in RAID configuration by allowing snapshots to be created with different RAID types while maintaining data consistency through the tree structure. The system can change RAID parameters for individual snapshots or groups of snapshots without compromising overall data consistency, resolving the contradiction between adaptability and reliability.
3Ease of operation
If mirror clones cannot be created when snapshots exist, then storage space is conserved, but operational flexibility is reduced
Solution Approach 1:
The patent separates snapshot metadata from actual snapshot data by using a snapshot tree and standalone tree structure. This segmentation enables the system to create mirror clones and snapclones independently of existing snapshots, as the clone creation process operates on the standalone tree rather than being blocked by the presence of snapshots in the snapshot tree, thus improving ease of operation while managing storage space efficiently.
4Quantity of substance
If snapshots are deleted to create space for new ones, then storage space is optimized, but information loss occurs
Solution Approach 1:
The patent divides snapshot management into a snapshot tree for metadata and a standalone tree for data. This segmentation allows the system to optimize storage space in the standalone tree without deleting snapshots from the snapshot tree, as the standalone tree can be managed independently to reclaim space while preserving all snapshot information in the metadata structure.
Solution Approach 2:
The patent enables the system to discard unused snapshot data in the standalone tree to optimize storage space while recovering and preserving snapshot metadata in the snapshot tree. This allows space optimization without information loss, as the metadata structure maintains complete snapshot information even when data is reclaimed.
Data Source
AI summary
Embodiments include methods, apparatus, and systems for managing snapshots in storage systems. One method includes logically linking a plurality of snapshots to a logical disk in a storage network; and splitting the snapshots from the logical disk to form a snapshot tree that includes the snapshots split from the logical disk.