File Location Map for Multi-Volume Storage Tiering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-volume file systems, the overhead of periodic relocation scans becomes significant as the number of files grows, especially when using solid-state disks (SSDs), due to the need for efficient SSD utilization and the impracticality of frequent scans in large file systems.
Innovation Solution
A file location map is generated and updated based on file creations and deletions, allowing for the identification of inactive files on high-performance SSDs, enabling their relocation to lower-performance tiers without extensive scanning, thus optimizing SSD usage and reducing resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If periodic relocation scans are performed to identify files for moving between storage tiers, then files can be relocated to optimize storage costs and performance, but the I/O and processing overhead increases significantly as the number of files grows into the millions
Solution Approach 1:
The system performs preliminary actions by maintaining a file location map that is continuously updated with file creation, deletion, and relocation information. This preliminary tracking eliminates the need for periodic full scans of the file system, as the map already contains current location information when relocation is needed.
Solution Approach 2:
The patent introduces an intermediary data structure called a file location map that mediates between the file system metadata and the relocation decision-making process. This map serves as a lightweight intermediary that can be quickly scanned and updated, avoiding the need to scan the entire file system hierarchy.
2Speed
If SSDs are used to store active files to improve performance, then I/O performance increases significantly, but the cost per byte increases by an order of magnitude and SSD endurance is limited
Solution Approach 1:
The system implements dynamic storage tiering that automatically adjusts file locations based on activity patterns. Files are dynamically moved between SSD and disk tiers based on observed access patterns, allowing the system to adapt to changing workload requirements and optimize the balance between performance and cost.
Solution Approach 2:
The patent changes the parameter of file location by monitoring file activity metrics and relocating files between storage tiers based on these changes. The system tracks file access patterns and uses these parameter changes to make intelligent decisions about which files should reside on SSD versus disk.
3Reliability
If multiple relocation scans per day are performed to optimize SSD utilization, then active files remain on SSD and inactive files are moved to disk, but resource consumption becomes impractical in file systems with large numbers of files
Solution Approach 1:
The file location map is preliminarily constructed and maintained through incremental updates as files are created, deleted, or relocated. This preliminary structure allows the system to quickly identify files on SSD without performing expensive full scans, enabling frequent optimization checks with minimal resource consumption.
Data Source
AI summary
A method for quickly identifying data residing on a volume in a multivolume file system. The method includes generating a file location map, the file location map containing a list of the locations of files that occupy space on each of a plurality of volumes of the file system. The file system comprises least a first volume and a second volume. The file location map is updated in accordance with changes in a file change log for the file system. Data residing on the first volume of the file system is identified by scanning the file location map.