Distributed File Locking via Access Layer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network file sharing systems lack effective control mechanisms to prevent simultaneous editing of shared files by local and remote users, leading to lost modifications and increased re-work, particularly due to limitations in Common Internet File Systems (CIFS) that do not support check-in/check-out file features and internet information services (IIS) based authentication.
Innovation Solution
A distributed access layer is introduced to manage network shares, allowing file handles with exclusive access permissions to be held across multiple nodes, enabling decentralized file locking and preventing write access to checked-out files, even by local users, while ensuring that remote users can edit files without conflicts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If CIFS protocols enable direct access to files by anyone with network access, then ease of operation is improved, but file access control and data integrity deteriorate
Solution Approach 1:
The patent introduces a distributed access layer as an intermediary between users and files. This layer implements check-in/check-out functionality that mediates file access requests, allowing direct access protocols to remain simple while adding control mechanisms. The access layer tracks file locks and prevents simultaneous edits without requiring changes to the underlying CIFS protocol.
Solution Approach 2:
The system segments file access control into separate components: the access layer handles locking logic while the file storage layer remains independent. This segmentation allows the access layer to enforce check-in/check-out policies without affecting the core file storage and retrieval operations, maintaining ease of operation while improving control.
2Reliability
If file handles are held to control exclusive access, then file access control is improved, but device complexity increases due to distributed node coordination
Solution Approach 1:
The patent extracts the file handle management and locking logic from individual nodes and concentrates it in the distributed access layer. This allows nodes to remain simple file storage units while the access layer handles the complexity of coordinating exclusive access across the distributed system, reducing per-node complexity while maintaining overall control.
Solution Approach 2:
The distributed access layer serves multiple functions: it manages file handles, enforces locking policies, tracks file state, and coordinates across nodes. This universal component handles all access control complexity centrally, allowing individual nodes to remain simple while the system as a whole achieves sophisticated distributed file locking.
3Reliability
If check-in check-out techniques are implemented, then file access control is improved, but adaptability to existing protocols deteriorates
Solution Approach 1:
The patent adds a new dimension to file access by introducing the access layer between users and files. This additional layer provides check-in/check-out functionality without modifying existing file protocols, allowing the system to support both traditional direct access and controlled access modes simultaneously, thereby improving adaptability.
4Reliability
If file locking information is decentralized across nodes, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent combines file locking information management with the distributed access layer, merging multiple functions (handle tracking, lock state management, node coordination) into a single unified component. This consolidation improves reliability through decentralization while managing complexity by providing a unified interface and consistent data model across all nodes.
Data Source
AI summary
Methods and systems for file locking are described herein. An on-premise file share may store files that are accessible to both a local on-premise client and a remote off-premise client. The off-premise file share may request to check-out one of the files. In response, one of multiple nodes may obtain for the file a file handle with exclusive write access. File locking information may be stored at the file share that indicates the node that holds the file handle and that indicates the file is in a locked state whereby other remote off-premise clients or local on-premise clients are prevented from editing the file.