Virtual File System FUSE Access Distributed Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing data storage needs of enterprises lead to high operating costs due to the costly configuration of multiple NAS devices, and accessing open-source distributed file systems through APIs requires complex application compilation, increasing labor costs.
Innovation Solution
The method involves creating a virtual file system in the operating system using FUSE, allowing access to open-source distributed file systems like HDFS or TFS without needing to access them through APIs, by mounting or unmounting network file systems as needed, thus mimicking local storage access modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple NAS devices are configured to meet growing data storage needs, then storage capacity is improved, but operating costs increase significantly
Solution Approach 1:
The patent combines multiple distributed storage devices (HDFS, TFS, Ceph) into a unified virtual file system that presents a single namespace to the operating system. This merging approach allows the system to utilize multiple storage devices for increased capacity while maintaining a simplified access interface, thereby avoiding the need to configure and manage multiple separate NAS devices and reducing associated operating costs.
Solution Approach 2:
The virtual file system implements multi-functionality by supporting access to multiple types of distributed storage systems (HDFS, TFS, Ceph) through a single unified interface. This universal access mechanism allows the system to leverage diverse storage backends for enhanced capacity while maintaining consistent access methods, eliminating the need for separate configuration of each storage device type.
2Adaptability or versatility
If open-source distributed file systems are accessed through APIs, then access flexibility is improved, but application compilation complexity increases
Solution Approach 1:
The patent introduces a virtual file system as an intermediary layer between the operating system and distributed storage devices. This intermediary translates standard file system operations into appropriate API calls for the underlying distributed storage system (HDFS, TFS, or Ceph), thereby maintaining access flexibility while shielding applications from compilation complexity. The virtual file system handles all API interactions, allowing applications to use simple file access commands without needing to understand or compile against complex distributed storage APIs.
Solution Approach 2:
The virtual file system creates a copy or abstraction of the distributed storage interface that mimics traditional local file system semantics. This copied interface presents familiar file operations (read, write, open, close) to applications, eliminating the need for applications to be recompiled or modified to work with distributed storage systems. The virtual file system layer handles the translation to the actual distributed storage API calls.
3Adaptability or versatility
If network file systems are mounted frequently to access different distributed storage devices, then storage versatility is improved, but system operation complexity increases
Solution Approach 1:
The patent merges multiple network file system mounts into a single virtual file system namespace. Instead of requiring separate mount operations for each distributed storage device (HDFS, TFS, Ceph), the virtual file system consolidates access to all these devices through one unified mount point. This merging eliminates the need for frequent mounting and unmounting operations while maintaining the ability to access diverse storage backends, thereby improving ease of operation while preserving storage versatility.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to an access method and apparatus, wherein the methodincludes: judging whether a first creation request for creating, in an operating system, a virtual file system corresponding to a first storage device is received; creating, in a preset path in the operating system, the virtual file system corresponding to the first storage device by using an FUSE in the operating system when the first creation request is received; judging whether a first access request for the preset path is received; and sending the first access request to the virtual file system when the first access request for the preset path is received, such that the virtual file system accesses the first storage device according to the first access request. The operating system can successfully access the first storage device by using an application for originally accessing a second storage device, and it is unnecessary for skilled persons to compile in advance, in the operation system, a great number of applications for accessing the first storage device through the API interface provided by the first storage device, thereby reducing labor costs.