Directory Copying via i-Node Tree Metadata Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for copying directories and file systems are slow and inefficient, relying on recursive copy operations that cause multiple operations over networks between client and server.
Innovation Solution
A method and system that parse a directory into an i-node tree, generate metadata, and transmit it to a target server to create a corresponding directory and copy files efficiently, using processors and firmware to manage the copying process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If recursive copy operations are used for each file component and contents, then the directory copying process is simple to implement, but the copying speed is slow and network operations are excessive
Solution Approach 1:
The patent segments the directory copying process into two distinct phases: metadata transmission phase and data transfer phase. The metadata (i-node tree structure, file attributes, directory hierarchy) is transmitted first to establish the target directory structure, followed by the actual file data transfer. This segmentation allows the system to prepare the destination structure in advance, enabling parallel or pipelined operations and reducing the total copying time by eliminating sequential dependencies.
Solution Approach 2:
The patent performs preliminary actions by transmitting and establishing the complete metadata (i-node tree) of the source directory structure before initiating the actual file data transfer. This preliminary metadata transmission creates the target directory framework in advance, allowing the subsequent data transfer to proceed more efficiently without repeated structure creation operations at the destination, thereby improving overall copying productivity.
2Productivity
If recursive copy operations are performed for each file, then individual file copying is straightforward, but multiple network operations between client and server are required
Solution Approach 1:
The patent merges multiple separate network operations into two consolidated network transmissions: one for the complete metadata (i-node tree) and another for the file data contents. Instead of performing individual copy operations for each file that would generate multiple network round-trips, the system combines all metadata operations into a single transmission and all data transfer operations into another transmission, significantly reducing the total number of network operations and improving network operation efficiency.
Solution Approach 2:
The patent uses metadata copying (i-node tree transmission) to replicate the directory structure information from source to destination before copying actual file data. This approach allows the target system to reconstruct the complete directory hierarchy and file metadata in advance, reducing the need for repeated structure creation and validation operations during the actual data transfer, thereby reducing overall network operation complexity.
3Productivity
If traditional file system copying methods are used, then compatibility with existing systems is maintained, but the copying process is inefficient and time-consuming
Solution Approach 1:
The patent changes the operational parameters of the copying process by introducing a two-phase approach with distinct protocols: first transmitting metadata with specific structure information (i-node tree) and then transferring data contents. This parameter change in the copying methodology enables more efficient data movement by optimizing the sequence and content of network transmissions, achieving higher copying efficiency while maintaining sufficient compatibility through standardized metadata formats.
Data Source
AI summary
A method, storage control system and information handling system (IHS) provides copying of a first directory containing a first plurality of files from a client to a target server. The method includes receiving, via a first processor of the client, a request to copy the first directory containing the first plurality of files from a first storage device to the target server. The first directory is parsed into a first i-node tree. Metadata corresponding to the first i-node tree is generated and transmit from the client to the target server. A second processor at the target server is triggered to generate a second directory based on the metadata corresponding to the first i-node tree. The first plurality of files contained in the first directory from the first storage device is transmitted to the target server.