File System Defragmentation During Over-Write
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
File fragmentation in digital computer systems leads to decreased read performance due to non-contiguous file system block allocations, which existing background defragmentation methods often contend with user I/O and metadata, making them inefficient.
Innovation Solution
A method that performs defragmentation during file over-write operations by determining fragmentation and allocating contiguous file system blocks, substituting them for existing allocations, thereby reducing fragmentation without introducing additional contention with user operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If flexible file system block allocation is used, then storage utilization efficiency is improved, but file fragmentation increases leading to decreased read performance
Solution Approach 1:
The patent performs defragmentation proactively during write operations before read operations occur. When a write operation is detected on a fragmented file, the system automatically consolidates the file's blocks into contiguous space, ensuring optimal read performance is maintained without requiring separate defragmentation passes.
Solution Approach 2:
The file system automatically detects and corrects its own fragmentation issues during normal write operations. The defragmentation process is triggered by write operations themselves, allowing the system to self-correct without external intervention or background processes that would contend with user I/O.
2Productivity
If background defragmentation process is run, then file fragmentation is reduced, but contention with user I/O and metadata access increases
Solution Approach 1:
The patent merges the defragmentation function with the existing write operation pathway. Instead of running separate background defragmentation processes, the defragmentation logic is integrated into the write path, so that writes and defragmentation occur in a unified, coordinated manner without competing for resources.
Solution Approach 2:
The useful action of writing data to the file system continues uninterrupted while defragmentation occurs. The write operation and defragmentation process work together continuously, with the write triggering and driving the defragmentation, eliminating idle background processes that would contend with user operations.
3Adaptability or versatility
If file blocks are allocated non-contiguously, then storage flexibility is improved, but data locality is reduced affecting cache efficiency
Solution Approach 1:
The patent applies local quality by making file blocks contiguous in physical storage space when they belong to the same logical file. This localized contiguity improves cache efficiency and data locality for that specific file region, while the overall file system maintains flexibility through its general allocation policies. The contiguity is applied locally to fragmented regions rather than forcing global restructuring.
Data Source
AI summary
When a file system manager responds to a request to write data to a specified number of blocks of a file, the file system manager determines that corresponding file system data block allocations exist and determines that the file is fragmented over the corresponding file system data block allocations. In response to determining that the file is fragmented over the corresponding file system data block allocations, the file system manager allocates and substitutes a single extent of contiguous file system data blocks for the corresponding file system data block allocations, and writes data to the newly allocated contiguous file system data blocks. For example, the file system manager compares a spread and a count of extent fragments of the file system data block allocations to appropriate thresholds to trigger the defragmentation upon file over-write.