Storage Control Device Snapshot Update Response Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The conventional copy-on-write method for creating snapshots in data backup systems results in prolonged response times due to the need to copy data prior to updates, which impairs response performance.
Innovation Solution
A storage control device and method that write update data to a snapshot volume instead of the original storage region, eliminating the need for data copying between storage regions during updates, and using bit maps to manage data access between snapshot and original volumes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the conventional copy-on-write method is used to create snapshots, then data backup functionality is achieved, but the response time for update requests is prolonged due to the need to copy data prior to updates
Solution Approach 1:
The patent segments the storage system into two distinct volumes: a first storage region for original data and a second storage region (snapshot volume) for snapshot data. By dividing the storage functionality, the system can maintain snapshot data without requiring copy operations during updates to the original storage region, thus resolving the contradiction between maintaining backup functionality and reducing update response time.
Solution Approach 2:
The patent introduces a bit map as an intermediary data structure that tracks the correspondence between logical addresses in the first storage region and the second storage region. This bit map enables the system to determine whether data exists in either volume without performing copy operations, thereby maintaining data backup reliability while eliminating the time-consuming copy process during updates.
2Reliability
If data is copied to snapshot storage region during updates, then snapshot integrity is maintained, but system performance deteriorates due to the copying overhead
Solution Approach 1:
The patent uses a logical copy approach rather than physical data copying. The bit map stores correspondence information between logical addresses in the first and second storage regions, allowing the system to maintain snapshot integrity through logical tracking rather than physical duplication. This eliminates the performance overhead of actual data copying while preserving snapshot integrity.
Solution Approach 2:
The patent changes the state tracking mechanism from physical data presence to logical address correspondence. By using a bit map to record which logical addresses have data in the snapshot volume versus the original volume, the system maintains snapshot integrity through parameter (bit map state) changes rather than through data copying operations, thereby improving system performance.
3Quantity of substance
If a shared logical volume is used for all snapshot replication volumes, then storage usage efficiency is improved, but access complexity increases
Solution Approach 1:
The patent makes the second storage region (snapshot volume) serve multiple functions: it acts as both a snapshot storage volume and a valid data storage volume. When data is written to the first storage region, the same logical address in the second storage region automatically receives the data, eliminating the need for separate copy operations. This multi-functionality improves storage efficiency while the bit map manages access complexity by providing clear lookup rules.
Data Source
AI summary
A storage control device includes a memory and a processor. The processor configured to, when a request for writing first data to a first unit region among unit regions included in a first storage region is received, write the first data to the first unit region in a second storage region corresponding to a snapshot, the snapshot being corresponding to the first storage region. The processor is further configured to send a first completion response to the request for writing. The processor is further configured to, when a first read request requesting for reading from the first unit region in the second storage region is received, send second data stored in the first unit region in the first storage region, as read data in accordance with the first read request.