Backup Management System for Networked Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computer systems where multiple PCs use networked storage devices, existing backup methods fail to detect when Logical Units (LUs) are in use, leading to incomplete or interrupted backup processes, as they initiate backups based on pre-designated times without ensuring the LUs are available.
Innovation Solution
A backup management system that monitors the status of LUs, inhibits writing to them during backup processes, and schedules backups based on designated frequencies, using volume replication and snapshots to ensure data integrity and availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the storage device starts the backup process at a designated date and time, then the backup can be performed automatically, but the backup cannot be completed if the LU is still being used with I/O activity
Solution Approach 1:
The storage device monitors the usage state of the LU and provides feedback to the backup management server. When the LU becomes available (unmounted or idle), the storage device notifies the backup management server, which then triggers the backup process. This feedback mechanism ensures the backup starts only when appropriate, resolving the contradiction between automatic execution and successful completion.
Solution Approach 2:
The backup execution timing is made dynamic rather than fixed. Instead of starting backup at a predetermined time regardless of LU state, the system adapts the backup start time based on the actual LU availability. The backup management server flexibly schedules the backup to begin when the LU transitions to an available state, ensuring both automation and reliability.
2Extent of automation
If the backup process is initiated at a pre-designated time, then automated backup scheduling is achieved, but data integrity cannot be ensured if the LU is in use
Solution Approach 1:
The system uses feedback from the storage device about LU usage state to determine the appropriate moment for backup. The backup management server receives notifications about when the LU becomes available and schedules the backup accordingly, ensuring data integrity while maintaining automated scheduling.
Solution Approach 2:
The system performs preliminary monitoring of the LU state before initiating the backup process. By checking whether the LU is available in advance and waiting for the appropriate condition, the system ensures that the backup will be performed on consistent data without interference from ongoing I/O operations.
3Reliability
If manual backup designation is required for each PC, then backup timing can be precisely controlled, but user burden increases and automation is reduced
Solution Approach 1:
The storage device performs self-service by automatically monitoring its own LU usage state and notifying the backup management server when appropriate for backup. This eliminates the need for users to manually check or designate backup timing for each PC, as the system autonomously determines and executes backups at appropriate moments.
Solution Approach 2:
The backup management server provides a universal solution that manages backups for multiple PCs and LUs centrally. Instead of requiring individual user actions for each PC, the single backup management server handles all backup scheduling and coordination, simplifying user operations while maintaining precise timing control through automated LU state monitoring.
4Reliability
If the storage device monitors LU usage state to ensure backup availability, then backup reliability improves, but system complexity increases
Solution Approach 1:
The backup management server acts as an intermediary between the storage device and the PCs. It centralizes the complexity of monitoring LU usage states and coordinating backups, while the storage device and PCs maintain relatively simple functions. This intermediary approach improves backup reliability through comprehensive monitoring without significantly increasing the complexity of individual components.
Data Source
AI summary
For a computer system that includes a storage system and a management computer, the storage system includes a first storage area, for storing data used by a computer connected to the storage system, and a second storage area, for storing a duplicate of the data included in the first storage area. The management computer includes first management information, for managing the frequency at which the data in the first storage area are copied. When the management computer receives a notification indicating the first storage area has been unmounted from another computer, the management computer employs the first management information for determining whether the data in the first storage area should be copied. Then, when the data should be copied, the first storage area is mounted on the computer and the data in the first storage area are copied to the second storage area.