Direct Image Lookup Metadata for Faster Snapshot Differentials
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of finding differentials between snapshots and linked targets is slow and inefficient, leading to inconsistencies in performance and impacting copy throughput, particularly in large-scale storage systems.
Innovation Solution
Implementing Direct Image Lookup (DIL) metadata to store images, allowing direct lookup of user data and comparing metadata pages of subsequent snapshots to identify changed tracks, thereby reducing computational resources required for differential processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional differential processing scans each track of an entire device to compare snapshots, then complete differential comparison is achieved, but processing speed decreases and copy throughput is impacted
Solution Approach 1:
The patent segments the storage device into tracks and organizes them into metadata pages. Instead of scanning all tracks sequentially, the system compares metadata pages to identify only the track sets that have changed between snapshots. This segmentation allows the differential process to focus only on relevant portions of data, significantly improving copy throughput while maintaining comparison accuracy.
Solution Approach 2:
The patent performs preliminary action by creating and maintaining metadata pages that index track locations and changes before the differential process runs. The metadata pages are updated during snapshot creation, so when differential processing is needed, the system already has organized information about which tracks have changed. This preliminary organization eliminates the need to scan entire devices during differential comparison.
2Reliability
If traditional differential processing scans all tracks regardless of changes, then complete data comparison is ensured, but computational resources are wasted on unchanged data
Solution Approach 1:
The patent extracts the essential comparison information into metadata pages, which contain only the track location and change status data. By taking out the detailed track data from the full device scan and placing it in compressed metadata pages, the system can perform differential processing on a much smaller data structure. This extraction maintains reliability by preserving all necessary comparison information while dramatically reducing computational resource consumption.
Solution Approach 2:
The patent creates a copy of track information in the metadata pages rather than working with the actual track data. The metadata pages are simplified representations that copy only the essential identification and change information. This copying approach allows the system to perform complete differential comparisons by comparing metadata page representations rather than scanning all actual track data, reducing energy consumption while maintaining completeness.
3Productivity
If metadata pages are compared to identify changed tracks, then processing speed improves, but system complexity increases
Solution Approach 1:
The patent makes the metadata pages serve multiple functions: they index track locations, track changes between snapshots, and provide the data structure for differential processing. By making the metadata pages multi-functional, the system avoids needing separate complex structures for each function. This universality improves processing speed by using a single efficient data structure while managing complexity through purposeful design rather than adding separate systems.
Data Source
AI summary
Direct Image Lookup (DIL) metadata is used to perform differential operation for snapshots and linked targets. Each snapshot contains metadata to store “images” to perform direct lookup of user data for any part of any snapshot in the system. Metadata pages of DIL image data represent sets of tracks of data. Metadata pages of subsequent snapshots for the same sets of tracks are compared, and where there are no changes to a given metadata page for a given set of tracks between subsequent snapshot copies, the differential process is not run on the tracks associated with the metadata page. If a given metadata page associated with a set of tracks has changed between subsequent snapshots, the differential process is used to identify which tracks in the subsequent snapshot contain different data than the corresponding tracks of the previous snapshot. Similar differential processing also is implemented for relinked target devices.