Segmented Dataset Backup for Tape Restore Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data backup methods, especially when transferring data from disk to tape, are inefficient due to the need for full incremental duplications, which complicate restores and prolong the process, whereas existing synthetic backup processes optimize disk backups but not tape duplications, leading to impractical restore times and excessive image requirements.
Innovation Solution
A method that divides a dataset into N segments, writes each segment sequentially to a second storage system, and includes both changed and unchanged data, allowing for efficient backups and restores by limiting the number of images needed for a full restore to N, thereby simplifying and shortening the restore process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional full incremental duplications are performed for tape backup, then data backup completeness is ensured, but restore complexity increases and restore time is prolonged
Solution Approach 1:
The backup image is divided into multiple segments, where each segment contains a portion of the original data plus forward incremental changes. This segmentation allows the system to store comprehensive backup data while enabling simplified restore operations by only requiring the most recent N segments rather than all incremental backups.
2Reliability
If traditional full incremental duplications are performed for tape backup, then data backup completeness is ensured, but the backup process time is prolonged
Solution Approach 1:
The system performs preliminary synthesis of backup images by pre-assembling segments that include both base data and forward incrementals before writing to tape. This preliminary action consolidates what would otherwise require multiple sequential incremental duplication operations into a single efficient backup process.
3Productivity
If synthetic backup processes are used to optimize disk backups, then disk backup efficiency is improved, but tape duplication remains inefficient requiring excessive images
Solution Approach 1:
The system merges the advantages of synthetic backups with efficient tape duplication by combining base images and forward incrementals into synthesized segments. This merging allows disk backup efficiency to be maintained while reducing the number of tape images from potentially dozens to just N segments, eliminating the excessive image requirement.
4Reliability
If traditional tape duplication methods are used, then data is backed up to tape, but operational efficiency is reduced and predictability is lost
Solution Approach 1:
The system changes key parameters of the tape backup process by limiting the number of required segments to a fixed value N, making backup window requirements predictable. This parameter change transforms an inefficient, unpredictable process into one with known, manageable resource requirements while maintaining complete data backup capability.
Data Source
AI summary
A method for managing a backup procedure is described. In one embodiment, the method includes performing a first backup of a dataset, selecting a value N, dividing the dataset into N segments, and upon writing at least the portion of the dataset to the second storage system, performing a first rewrite. Each segment of the dataset includes 1/Nth of the dataset. The first backup includes writing at least a portion of the dataset from a first storage system to a second storage system. The first rewrite includes writing a first segment of the dataset from the first storage system to the second storage system.