Filesystem Block Sampling for Storage Resource Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current file systems lack efficient identification and management of users who overconsume storage resources, leading to inefficient data backup and archiving processes, as they do not quickly identify unneeded files for deletion or archiving, which affects storage efficiency and costs.
Innovation Solution
A file system that employs statistical analysis through random sampling of data blocks to identify individual user consumption of storage resources, using cylinder groups with reverse mapping data structures to track file allocations and permissions, enabling efficient identification of files for deletion or archiving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional file systems are used to store large amounts of data, then storage capacity is increased, but the ability to efficiently identify users who overconsume storage resources deteriorates
Solution Approach 1:
The patent segments the filesystem into cylinder groups, each containing multiple data blocks. This segmentation allows the system to sample and analyze storage consumption in manageable units rather than processing the entire filesystem at once, enabling efficient identification of overconsuming users while maintaining support for large storage capacities.
Solution Approach 2:
The patent employs statistical sampling of data blocks within cylinder groups rather than examining every single block. This partial action approach provides sufficiently accurate identification of storage consumption patterns while dramatically reducing the computational effort required, thus improving identification efficiency without sacrificing storage capacity.
2Reliability
If data backup mechanisms employ mirroring and replication, then data reliability is improved, but storage costs and time consumption increase
Solution Approach 1:
The patent performs preliminary identification of overconsuming users and their files before backup operations are executed. By using cylinder group sampling to pre-identify files that should be deleted or archived, the system prepares the data set for backup in advance, reducing the actual backup time and allowing mirroring and replication to operate on a smaller, more manageable data set.
3Measurement precision
If comprehensive file tracking is implemented, then storage management accuracy is improved, but system complexity increases
Solution Approach 1:
The patent achieves sufficient storage management accuracy through statistical sampling of cylinder groups rather than comprehensive tracking of every data block. This partial monitoring approach provides accurate enough information to identify overconsuming users and their files while avoiding the excessive complexity that would result from complete filesystem analysis.
Solution Approach 2:
By dividing the filesystem into cylinder groups and sampling within those segments, the patent achieves manageable system complexity. The segmentation allows the use of reverse mapping data structures that efficiently track file allocations at the cylinder group level without requiring complex global tracking mechanisms.
Data Source
AI summary
Providing a statistical analysis of all files in a file system based on random sampling of data blocks to identify individual user consumption of file system resources and characteristics of the files stored in the file system. In one or more of the various embodiments, the file system is based on information for a plurality of cylinder groups. Also, each cylinder group may include at one or more known locations at least three types of data structures that enable reverse mapping of data blocks to root directories.