Dynamic Writer Discovery for Incremental Backup Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current backup systems face challenges in efficiently performing incremental backups of bare metal recovery (BMR) and system state recovery (SSR) objects due to complexities and interdependencies, leading to increased storage costs and backup times, as well as limitations in accurately reporting system asset sizes and user experience issues during restoration.
Innovation Solution
The implementation of automated backup recovery techniques in a distributed system, utilizing a data protection backup system that includes a management console, backup catalog, and recovery manager to conduct incremental backups by identifying and tagging writers, creating base backup copies, and dynamically promoting user data writers to system writers for comprehensive backup and recovery operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If incremental backup is not supported for BMR/SSR objects, then backup time and storage costs increase due to full backups being required each time, but implementing incremental backup requires overcoming the complexity and interdependencies of system data
Solution Approach 1:
The patent segments system data into distinct components (system state data, boot volume, system volume) and introduces writer components that can independently track and backup each segment. This allows incremental backups by monitoring changes in individual segments rather than treating the entire system data as a monolithic unit, thereby resolving the contradiction between backup efficiency and system complexity.
Solution Approach 2:
The patent introduces writer components as intermediary elements between the backup system and system data. These writers act as mediators that monitor, track, and report changes in system data components, enabling the backup system to perform incremental backups without directly dealing with the underlying complexity and interdependencies of system data.
2Adaptability or versatility
If new system components are added in OS releases, then backup coverage improves, but existing backup products cannot accurately report asset sizes and may miss new components
Solution Approach 1:
The patent implements a dynamic discovery mechanism where the backup system continuously monitors and discovers new system components, writers, and data assets. Rather than relying on static inventories, the system dynamically adapts to new OS components and accurately reports asset sizes by querying the actual state of system data, thereby improving both backup coverage and measurement precision simultaneously.
Solution Approach 2:
The patent incorporates feedback mechanisms where writer components continuously report the state and size of system data they manage. This feedback loop enables the backup system to accurately report asset sizes and detect new components added in OS releases, as the writers provide real-time information about their associated data without requiring manual configuration or static inventories.
3Ease of operation
If manual network configuration is required during BMR restore, then restore process can be customized, but user experience deteriorates and restore time increases
Solution Approach 1:
The patent applies preliminary action by automatically performing network configuration during the BMR restore process before the user needs to use the system. The backup system pre-configures network settings based on the restored system's requirements, eliminating the need for users to manually recall and configure network parameters, thereby reducing restore time and improving ease of operation.
Solution Approach 2:
The patent implements self-service where the BMR restore process automatically configures its own network settings without requiring user intervention. The system retrieves and applies network configuration information autonomously during restoration, allowing the restore process to serve itself rather than relying on manual user configuration, thus reducing both restore time and operational complexity.
Data Source
AI summary
A data structure is generated to store information about a set of writers associated with a copy service. The information is gathered from the copy service and populated into the data structure. The information includes names of the writers, writer types as being system or application writers, and system writer subtypes as being one of bare metal recovery (BMR) writers or system state recovery (SSR) writers. The data structure is reviewed to determine whether a particular writer should participate in the backup. Based on the review, a list is generated including a subset of the participating writers. Instructions are issued, via the copy service, for the writers in the list to prepare for creation of a snapshot. Upon creation of the snapshot, data is backed up from the snapshot to secondary storage.