Mirror Manager Application Program I/O Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computer storage systems with mirrored volumes, data inconsistency and prolonged degraded mode operations occur due to incomplete write requests during system crashes, leading to significant performance degradation during recovery processes, especially in large systems with many mirror volumes.
Innovation Solution
The implementation of an active map and recovery map by the Mirror Manager Application Program (MMAP) to track and manage write requests, ensuring data consistency by maintaining an active map in memory and writing it to disk, and using a recovery map to synchronize data across mirrored volumes, allowing for intelligent processing of I/O requests and reducing the need for dirty logs and clear logs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If parallel recovery operations are scheduled for multiple mirror volumes, then recovery speed is improved, but system complexity and resource management difficulty increase
Solution Approach 1:
The patent segments the mirror volume recovery process into independent parallel operations, where each mirror volume can be recovered simultaneously without interfering with others. The system divides the recovery workload across multiple volumes, allowing concurrent processing while maintaining manageable complexity through structured resource allocation.
Solution Approach 2:
The system dynamically adjusts the number of parallel recovery operations based on available system resources and workload conditions. It can flexibly scale the degree of parallelism, increasing recovery speed when resources are abundant while reducing complexity when system load is high, achieving optimal balance between speed and manageability.
2Reliability
If the system operates in degraded mode for extended periods to complete recovery, then data consistency is ensured, but I/O performance deteriorates
Solution Approach 1:
The system performs preliminary actions by proactively identifying and initiating recovery operations before they are critically needed. By starting recovery processes early and maintaining them in progress, the system reduces the duration of degraded mode operation, thereby minimizing I/O performance impact while ensuring data consistency is restored promptly.
Solution Approach 2:
The patent maintains continuous recovery operations in the background while allowing normal I/O operations to continue on healthy mirror volumes. This continuity ensures that recovery progresses steadily without interrupting available storage capacity, maintaining both data consistency and acceptable I/O performance throughout the recovery process.
Data Source
AI summary
A method, apparatus and software is disclosed for processing input/output (I/O) requests for a mirrored storage volume in recovery mode in which the processing of normal I/O is optimised using the recovery map for the volume.