Custom Namespace File System Plug-in Engine
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage systems face inefficiencies due to client-side processing of operations on logical constructs, which increases network usage and load on the storage system, as clients perform multiple requests for post-processing and metadata operations.
Innovation Solution
A custom namespace is introduced within the pathname of file system protocol requests, allowing a 'plug-in' engine on the storage system to semantically interpret commands and perform predefined operations, thereby offloading these tasks from the client to the storage system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the client performs post-processing operations and metadata operations locally, then the client can access and process files, but network usage increases and the storage system load increases due to multiple requests
Solution Approach 1:
The patent combines multiple separate file access requests into a single gang operation request. Instead of the client making individual requests for each file operation (open, read, close, metadata operations), the system merges these into one consolidated request that processes multiple files simultaneously on the storage system, reducing network bandwidth consumption while maintaining client-side processing capability.
Solution Approach 2:
The patent introduces a new dimension of operation by implementing gang operations that process multiple files in parallel on the storage system side. This shifts the processing paradigm from sequential client-side operations to parallel server-side batch operations, effectively moving compute tasks from the client to the storage system and reducing network traffic.
2Adaptability or versatility
If the client performs recursive metadata operations and filtering, then the client can find files with specific criteria, but the time for performing operations increases and the storage system load increases
Solution Approach 1:
The patent applies preliminary action by performing metadata filtering and pattern matching operations on the storage system before files are transferred to the client. The storage system pre-processes metadata, filters files based on criteria (extensions, pattern matches), and prepares only the relevant results for transfer, significantly reducing operation execution time compared to client-side recursive filtering.
Solution Approach 2:
The patent extracts the computationally intensive metadata filtering and pattern matching operations from the client side and relocates them to the storage system side. By taking out these operations and executing them on the server, the system reduces the time required for metadata operations while maintaining the adaptability of filtering capabilities.
3Productivity
If the client issues multiple data access protocol requests for file operations, then the client can perform complex file processing, but the load on the storage system increases
Solution Approach 1:
The patent merges multiple data access protocol requests into a single gang operation request. Instead of the client issuing separate NFS/SMB requests for each file operation (open, read, write, close, stat), the system consolidates these into one request that handles multiple files and operations simultaneously, maintaining full file processing capability while reducing protocol request complexity.
Solution Approach 2:
The patent implements a universal gang operation mechanism that can handle multiple types of file operations (read, write, delete, metadata operations) across multiple files within a single request. This multi-functional approach maintains the productivity of complex file processing while reducing the overall complexity of protocol requests by making the single request type versatile enough to handle various operations.
Data Source
AI summary
A technique extends a file system infrastructure of a storage system to provide a custom namespace within a pathname of a logical construct configured to invoke semantically interpretative context as a command embedded in a data access protocol request issued by a client and directed to the logical construct served by the storage system, without alteration to the data access protocol. The extension includes a “plug-in” engine of a data access protocol server executing on a network protocol stack of the storage system. The engine operates to extract a pathname from the request to determine whether the custom namespace incorporating the command is present and directed to the logical construct. If so, the engine semantically interprets the command within a context of the custom namespace to essentially convert the command to one or more predefined operations directed to the logical construct. The storage system then performs the operations and returns the results to the client.