Quota Tree Inode Grouped Data Structure for File System Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional file systems face conflicts and limitations in maintaining and restoring quota trees due to shared inode identifiers, which hinder efficient copying and restoration processes, particularly in scenarios like logical replication and backup operations.
Innovation Solution
Implementing an inode grouped data structure for each quota tree, allowing each quota tree to maintain its own index of inode identifiers, thereby enabling identical inode identifiers across copies and eliminating the need for correlation information, supporting fan-in and cascade copies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single common table is used to store metadata for all quota trees, then device complexity is reduced, but inode identifier conflicts arise when copying quota trees between volumes
Solution Approach 1:
The patent divides the single common metadata table into separate inode grouped data structures for each quota tree. This segmentation allows each quota tree to maintain its own inode identifiers independently, eliminating conflicts when copying quota trees between volumes while preserving the simplified structure within each isolated group.
2Reliability
If inode identifiers are renumbered when copying quota trees to different volumes, then inode identifier conflicts are avoided, but restoration complexity increases due to need for correlation information
Solution Approach 1:
The patent enables direct copying of inode identifiers from source to destination quota trees without renumbering. By isolating inode identifier spaces per quota tree through separate inode grouped data structures, the system allows exact copies to be made, simplifying restoration processes and eliminating the need for correlation information tracking.
3Manufacturing precision
If correlation information is stored to track inode identifier mappings, then restoration accuracy is improved, but device complexity and storage requirements increase
Solution Approach 1:
The patent extracts the inode identifier management responsibility from the global namespace and places it within each individual quota tree's inode grouped data structure. This extraction eliminates the need for separate correlation information storage, as each quota tree's inode identifiers are self-contained and can be copied directly without tracking mappings.
4Quantity of substance
If a single volume is used as backup destination for multiple quota trees, then storage efficiency is improved, but inode identifier conflicts prevent exact copies from being created
Solution Approach 1:
The patent segments the inode identifier namespace by creating separate inode grouped data structures for each quota tree, even when they coexist on the same volume. This segmentation allows multiple quota trees to share a volume for backup purposes while maintaining their own independent inode identifier spaces, enabling creation of exact copies without conflicts.
Data Source
AI summary
A method of identifying metadata referenced by a quota tree. A quota tree metafile is accessed, and this quota tree metafile includes references to locations of inode grouped data structures associated with quota trees. Here, each quota tree is allocated an inode grouped data structure. A reference to a location of an inode grouped data structure that is allocated to the quota tree is identified from the quota tree metafile. The inode grouped data structure is accessed based on the reference, and this inode grouped data structure defines a set of inode identifiers associated with the quota tree. An inode grouped data structure that stores the metadata is then located using the set of inode identifiers as index.