Intelligent Backup Model for Snapshot Cataloging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current backup systems face challenges during peak operating times as they divert computing resources away from customer requests and production activities, leading to long wait times and inefficiencies, even with advancements in snapshot technologies, due to the time-consuming nature of backup processes beyond snapshot creation.
Innovation Solution
The system implements intelligent backup initiation and cataloging processes, using a scheduler, object mapping, and cataloging policies to optimize resource usage by postponing cataloging to non-peak hours, reducing the need for frequent connections and allowing more snapshots to be taken, thereby improving RPO and RTO while offloading processing overheads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional backup processes are used during peak operating times, then data backup is performed, but computing resources are diverted from customer requests and production activities, leading to long wait times and reduced productivity
Solution Approach 1:
The system creates snapshots of data at scheduled intervals during peak hours as a preliminary action. These snapshots capture the state of data at specific points in time without requiring full backup processing during critical periods. The actual backup and cataloging operations are then performed on these pre-created snapshots during non-peak hours, ensuring both data protection and continuous high-performance customer service.
2Reliability
If frequent backups are performed to reduce data loss risk, then data protection is improved, but computing resources are continuously consumed, causing system slowdowns and increased operational costs
Solution Approach 1:
The backup process is segmented into distinct phases: snapshot creation during peak hours and backup/cataloging operations during non-peak hours. This segmentation allows the system to perform frequent snapshots for data protection while concentrating resource-intensive operations during periods of lower demand, thereby reducing overall computing resource consumption and avoiding system slowdowns.
3Measurement precision
If full cataloging is performed for every backup to ensure complete data tracking, then backup cataloging accuracy is improved, but the time and resources required for each backup operation increase significantly
Solution Approach 1:
Instead of performing complete cataloging for every snapshot, the system applies partial cataloging strategies. Cataloging is performed selectively based on change detection, data priority, and resource availability. This partial action approach maintains sufficient cataloging accuracy for recovery operations while dramatically reducing the time and resources required compared to full cataloging of every backup.
Data Source
AI summary
A discovery is made of an object to be backed up on a storage array of a client. An object mapping is generated that maps the object to a LUN of the storage array where the object is located. The object mapping is stored at a backup server. A snapshot of the storage array is obtained. The object mapping is reviewed to identify on the snapshot the object to be backed up and the object is backed up to a backup storage unit.