Unified File Object Storage Architecture Inode Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object storage systems face scalability issues due to growing container and account databases, unpredictable metadata update and retrieval times, replication challenges across multiple sites, lack of automatic metadata creation for objects generated via file interfaces, and incompatibility between file and object access control lists.
Innovation Solution
The implementation of an extended inode space in storage systems that allows for recursive metadata attribute storage and update, enabling automatic metadata generation and ACL compatibility across file and object interfaces, eliminating the need for container and account databases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If container and account databases are used to store metadata in object storage systems, then metadata management is enabled, but system scalability is limited due to database size growth
Solution Approach 1:
The patent extracts the metadata storage function from traditional container/account databases and relocates it to the file system inode structure. By storing metadata directly in inodes, the system eliminates the need for separate database layers, thereby resolving the scalability limitation imposed by growing database sizes while maintaining comprehensive metadata management capabilities.
Solution Approach 2:
The patent makes the inode structure multi-functional by having it serve both traditional file system purposes and object storage metadata storage. The inode becomes a universal structure that handles both file operations and object metadata management, eliminating the need for separate database systems and enabling unlimited object storage capacity.
2Reliability
If container and account databases are used for metadata storage, then object metadata can be managed, but metadata update and retrieval times become unpredictable under load
Solution Approach 1:
The patent removes the database layer from the metadata storage path and extracts metadata directly into the file system inode structure. This eliminates sequential database operations and enables direct, predictable metadata access through the file system, thereby reducing metadata retrieval time and improving update consistency under load conditions.
3Reliability
If database files are replicated across multiple sites for consistency and error recovery, then data reliability is improved, but replication time increases and database listings become inaccurate
Solution Approach 1:
The patent extracts metadata from centralized database files and stores it directly in the distributed file system inodes. This eliminates the need for database file replication across multiple sites, as each node can directly access and read metadata from the file system itself, thereby achieving data consistency without the time overhead and accuracy issues of database replication.
4Ease of operation
If objects are created via file interfaces, then file access is enabled, but automatic object metadata creation is missing
Solution Approach 1:
The patent merges the file interface creation process with object metadata generation by having the inode structure automatically capture and store metadata when files are created through file system operations. This integration ensures that objects created via file interfaces automatically receive proper metadata without requiring separate object interface operations, thereby enabling full automation.
Solution Approach 2:
The patent enables the file system itself to automatically generate and store object metadata through the inode structure during file creation operations. The system serves itself by automatically creating the metadata infrastructure as part of the file creation process, eliminating the need for separate manual metadata creation steps.
5Adaptability or versatility
If traditional object storage architecture is used, then object interface access is provided, but file and object ACL compatibility is not maintained
Solution Approach 1:
The patent makes the inode structure multi-functional by having it store both traditional file system permissions and object storage ACL information. This unified approach allows the same inode to serve both file and object access control requirements, thereby achieving compatibility between file and object interfaces while simplifying ACL management through a single unified structure.
Data Source
AI summary
A processor may identify a first directory in the storage system. The first directory may include one or more subdirectories in one or more levels under the first directory. The one or more subdirectories may include a second directory that has includes one or more objects. The first directory may be associated with a first inode, and the second directory may be associated with a second inode. The processor may perform a stat call on the second directory to determine metadata attributes for the one or more objects that are stored in the second directory. The metadata attributes for the one or more objects may be stored in the second inode. The processor may add the metadata attributes for the one or more objects to the first inode.