Object Store Inode Merging for Fast Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing object store systems face performance bottlenecks in data retrieval due to time-consuming mapping from object IDs to file path names, especially in cloud environments with large numbers of objects, where current methods either consume excessive memory or are computationally intensive.
Innovation Solution
Enhancing the object ID generation and mapping by merging the object ID with the inode ID and modifying the inode to include the object ID, allowing direct access and reducing the need for additional mapping tables, thus optimizing resource utilization and improving retrieval performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If mapping from object ID to file path name is performed using in-memory copy, then object retrieval speed is improved, but memory consumption becomes impractical for large numbers of objects
Solution Approach 1:
The patent extracts the mapping functionality from centralized in-memory storage and distributes it to individual inodes in the file system. Each inode contains its own object ID to file path mapping, eliminating the need for a separate centralized mapping table and reducing memory consumption while maintaining fast access.
Solution Approach 2:
The patent introduces the inode as an intermediary structure that bridges object ID and file path. Instead of directly maintaining object ID to file path mappings in memory, the system uses inodes as intermediaries that store this mapping information persistently on disk, reducing memory requirements while enabling fast retrieval.
2Quantity of substance
If database is used to store object ID to file path mapping, then memory consumption is reduced, but computational intensity increases due to searching through huge lists of object IDs
Solution Approach 1:
The patent segments the centralized mapping database into distributed mappings stored in individual inodes. Instead of searching through a huge centralized list of object IDs in a database, the mapping information is segmented and stored with each file's inode, enabling direct access without iterative searching and reducing computational intensity.
Solution Approach 2:
The patent performs preliminary action by pre-computing and storing the object ID to file path mapping information directly in the inodes during file creation. This eliminates the need for computational searching during object retrieval, as the mapping is already prepared and stored in the appropriate location.
3Ease of operation
If object ID to file path mapping is performed through centralized indexing, then object retrieval is simplified, but system performance deteriorates due to time-consuming mapping operations
Solution Approach 1:
The patent merges the object ID to file path mapping functionality with the inode structure itself. Instead of maintaining a separate centralized indexing system that requires additional lookup operations, the mapping information is combined with the file metadata in the inode, eliminating separate mapping operations and reducing time loss.
Data Source
AI summary
In one embodiment, a method for creating an object includes creating an object in an object store making use of a file system, obtaining an index node (inode) identifier (ID) for an inode associated with the object, creating an object ID for the object, wherein the object ID is unique from any other object IDs in the object store, merging the object ID with the inode ID, and modifying the inode to include at least the object ID. In another embodiment, a system includes a processor and logic integrated with and/or executable by the processor, the logic being configured to: create an object in an object store making use of a file system, obtain an inode ID for an inode associated with the object, create an object ID for the object, the object ID including the inode and modify the inode to include at least the object ID.