Backup Transfer ETA Using Changed-Byte History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data transfer time estimation methods for backup jobs are inefficient due to the lengthy assessment mode, which impacts the overall completion time of the job.
Innovation Solution
Implementing Total Bytes To Transfer (TBTT) capabilities by reading and utilizing historical data transfer times to estimate the completion time of backup jobs, updating the TBTT in a database, and using mean transfer times to calculate an estimated time of arrival (ETA) dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system performs a complete assessment mode to estimate backup completion time, then the estimation accuracy is improved, but the overall job completion time increases
Solution Approach 1:
The system performs assessment mode in advance during backup job creation to pre-calculate the total bytes to transfer (TBTT). This preliminary action stores the assessment results in a database, eliminating the need to repeat the full assessment during backup execution, thus improving estimation accuracy without extending the actual backup time
Solution Approach 2:
The patent extracts only the essential assessment information (TBTT) needed for estimation and separates it from the complete assessment mode process. By storing just this critical parameter in the database, the system achieves accurate estimation without requiring the full assessment mode to run again during backup, reducing time loss while maintaining precision
2Productivity
If the system uses historical data transfer times to estimate completion time, then the estimation speed is improved, but the adaptability to changing transfer conditions deteriorates
Solution Approach 1:
The system dynamically updates the mean time to transfer a byte during backup execution by continuously monitoring actual transfer performance. This dynamic adjustment allows the estimation to adapt to changing transfer conditions while maintaining high estimation speed, resolving the contradiction between using historical data and adapting to new conditions
Solution Approach 2:
The patent implements feedback mechanisms where actual backup performance data is fed back into the estimation model. The system uses this feedback to recalculate the mean transfer time and update the estimated time of completion, ensuring adaptability to changing conditions while maintaining fast estimation through the use of pre-calculated TBTT
Data Source
AI summary
A system can maintain respective numbers of changed bytes for respective backup paths of a file system, which indicate how many bytes have changed since respective most-recent backups of the respective backup paths. The system can, while performing a backup of a backup path of the respective backup paths, for respective tasks of the backup, determine respective amounts of time spent backing up respective files associated with the respective tasks, and respective file sizes of the respective files, determine a mean time to transfer a byte based on the respective amounts of time spent backing up the respective files, and the respective file sizes, determine an estimated time of completing the backup based on the mean time to transfer a byte and the number of changed bytes for the backup path, and store an indication of the estimated time of completing the backup.