Network Caching Device Disk Fragmentation Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network caching appliances experience performance degradation due to disk fragmentation as storage space fills up, leading to inefficient write head movement and repositioning, especially when content is replaced haphazardly across physical disks.
Innovation Solution
Dividing storage space into physically contiguous stripes on hard disks, where content is written one stripe at a time, and when full, the oldest stripe is recycled as a whole, with cache hits rewritten to the currently written stripe, minimizing fragmentation and write head movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If content is replaced haphazardly on physical disks as storage fills up, then storage space utilization is maintained, but write performance degrades due to frequent write head repositioning
Solution Approach 1:
The patent divides the disk storage into multiple contiguous stripes, where each stripe is a physically contiguous section of the disk. Content is written to stripes sequentially rather than scattered randomly, which segments the write operations into organized blocks. This segmentation prevents fragmentation by ensuring that writes occur in contiguous regions, thereby maintaining write performance while filling storage space efficiently.
2Quantity of substance
If haphazard content replacement is used to manage storage space, then storage capacity is utilized, but write head movement increases causing performance degradation
Solution Approach 1:
The patent implements a preliminary action by pre-dividing the disk into stripes and establishing a systematic replacement strategy before storage fills up. Instead of reacting to storage fullness with random replacement, the system proactively manages stripe recycling in a predetermined order. This preliminary organization ensures that when storage is full, the replacement process follows a planned path that minimizes write head movement and maintains performance.
3Object-generated harmful factors
If entire stripes are recycled as a whole when full, then fragmentation is reduced, but the complexity of stripe management increases
Solution Approach 1:
The patent implements self-service by enabling the disk storage system to automatically manage its own stripe recycling without external intervention. The system autonomously tracks which stripes are full, identifies the oldest stripes for recycling, and performs the recycling process automatically. This self-service mechanism reduces fragmentation while managing complexity internally, eliminating the need for manual stripe management or complex external control systems.
Data Source
AI summary
Storage space on one or more hard disks of a network caching appliance is divided into a plurality S of stripes. Each stripe is a physically contiguous section of the disk(s), and is made up of a plurality of sectors. Content, whether in the form of objects or otherwise (e.g., byte-cache stream information), is written to the stripes one at a time, and when the entire storage space has been written the stripes are recycled as a whole, one at a time. In the event of a cache hit, if the subject content is stored on an oldest D ones of the stripes, the subject content is rewritten to a currently written stripe, where 1≦D≦(S−1).