Distributed File System Object Identifier Decoupling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing file systems require fully qualified paths to be stored on the same physical device, limiting data accessibility and flexibility, as they rely on attributes like storage location and device context, making it difficult to move or modify files without updating directory structures.
Innovation Solution
A distributed file system uses globally-unique object identifiers independent of file attributes, allowing data files to be universally accessible and stored across multiple nodes, eliminating the need for fully qualified paths by dynamically resolving storage locations based on object identifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fully qualified paths are used to map data objects to storage locations, then data objects can be organized in a hierarchical directory structure, but data objects must be stored on the same physical device as their directory structure
Solution Approach 1:
The patent segments the file system into two independent components: the namespace hierarchy (directories and paths) and the data object storage. By using globally unique object identifiers instead of path-based location mapping, the directory structure and physical storage location are decoupled, allowing data objects to be stored on any device regardless of where their parent directories reside.
Solution Approach 2:
The patent introduces a name service as an intermediary component that resolves globally unique object identifiers to actual storage locations. This mediator allows clients to access data objects using simple identifiers without needing to know or manage the complex physical distribution of files across the network, thus enabling flexible storage while maintaining simple access semantics.
2Adaptability or versatility
If data objects are moved to different physical devices, then storage flexibility and load balancing are improved, but directory structures and path mappings must be updated
Solution Approach 1:
The patent extracts the location information from the directory structure itself and stores it separately in the name service. When data objects are moved, only the name service needs to be updated with the new location mapping, while the directory structures remain unchanged. This separation eliminates the need to update directory entries during data object migration.
Solution Approach 2:
The name service automatically manages the mapping between object identifiers and storage locations, providing self-service location resolution. When a client needs to access a data object, the name service dynamically resolves the current location based on the globally unique identifier, eliminating the need for manual directory updates and reducing administrative overhead.
3Adaptability or versatility
If hard links are used to create additional access paths to data objects, then data accessibility is improved, but all hard links must remain on the same physical device and partition
Solution Approach 1:
The patent implements a universal naming mechanism where globally unique object identifiers can resolve to data objects regardless of their physical location. This universal identifier system replaces the device-specific hard link mechanism, allowing any node in the distributed file system to create access paths to any data object on any device, thus achieving multi-device accessibility with the same simplicity as traditional hard links.
4Adaptability or versatility
If mount commands are used to place device hierarchies at arbitrary namespace locations, then data accessibility is improved, but the entire tree under the mount point is restricted to the same device
Solution Approach 1:
The patent adds a new dimension to file system organization by introducing globally unique object identifiers that operate independently of the traditional path-based hierarchy. Instead of being constrained to device-specific mount points, data objects can be accessed through their universal identifiers from any location in the namespace, effectively adding a logical dimension that transcends physical device boundaries.
Data Source
AI summary
Each data file of a distributed file system is identified by a corresponding globally-unique object identifier that is independent of any attribute of the data file. A node in the distributed file system has a file system that includes a plurality of directory entries, where each directory entry has a file name field, an attributes field, and an object identifier field configured for specifying a globally-unique object identifier. The globally-unique object identifier is universally reachable by any node of the distributed file system and uniquely identifies the data file, enabling the data file to be universally identifiable by any node based on the corresponding object identifier. The data file can be stored independently of a device file system based on providing a resolution between the data file name and its object identifier, followed by a resolution between the object identifier and a selected location for the data file.