Dynamic Write Balancing via Non-Sequential Free Block Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional block re-use schemes in dynamic filesystems lead to imbalanced write operations due to sequential searching for free blocks, impacting filesystem performance, especially in large-scale storage systems with multiple storage devices.
Innovation Solution
Implementing non-sequential searching methods for dynamic write balancing, where the file server logically divides the free blocks object into sections and searches through them in a non-linear pattern across multiple stripesets or sections, distributing write operations across storage space to balance disk utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If sequential searching is used to locate free blocks, then the searching process is simple to implement, but write operations become imbalanced and filesystem performance deteriorates
Solution Approach 1:
The patent segments the storage space into multiple stripesets and divides the free blocks object into sections corresponding to different stripesets. Instead of sequentially searching through all free blocks in a single list, the system maintains separate free blocks lists or sections for each stripeset and searches through these segments in a round-robin or load-balanced manner. This segmentation allows write operations to be distributed across multiple stripesets simultaneously, preventing any single stripe from becoming a bottleneck while maintaining a relatively simple search mechanism within each segment.
2Device complexity
If sequential searching is used through filesystem storage space, then the search mechanism is straightforward, but disk utilization becomes imbalanced
Solution Approach 1:
The patent implements a dynamic block allocation strategy where the system continuously monitors and adjusts the distribution of write operations across stripesets. The free blocks object is dynamically organized into sections that correspond to different stripesets, and the system adapts the search pattern based on current disk utilization levels. When certain stripesets become more full, the system automatically directs new write operations to less-utilized stripesets, maintaining balanced disk utilization without requiring a complex static search mechanism.
3Ease of manufacture
If traditional block re-use schemes are used, then the implementation is simple, but performance bottlenecks occur in large-scale storage systems
Solution Approach 1:
The patent introduces a new dimension to block allocation by organizing free blocks not just as a single sequential list but as a multi-dimensional structure associated with stripesets. Each stripeset has its own section in the free blocks object, creating a two-dimensional organization (stripeset dimension and block position dimension). This allows the system to perform parallel or interleaved searches across multiple stripeset sections simultaneously, dramatically improving write operation speed in large-scale storage systems while maintaining implementation simplicity through the systematic organization of free blocks by stripeset.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Available storage locations are identified by searching through a free blocks object non-sequentially. The free blocks object may be logically divided into sections, and the sections may be searched non-sequentially. Each section may be logically divided into a number of sub-sections, and the sub-sections of the various sections may be searched non-sequentially. A new section may be added to the free blocks object, and the added section may be searched until a predetermined threshold or condition is reached, at which point non-sequential searching may begin or resume.