Mirror Destination Server Bypasses Filesystem Logic for Direct Write
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage protection systems face delays and increased complexity due to synchronous mirroring, where delays at one system can cause cascading issues at another, and existing backup methods are time-consuming and do not protect new data modifications between backups.
Innovation Solution
A mirror destination storage server maintains an identical copy of the mirror source filesystem, allowing data to be written directly to a hard disk or container file even when the filesystem is frozen, decoupling source and destination server operations without additional processing costs or reducing data protection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous mirror operations are implemented to maintain real-time data copies, then data protection is improved, but system delays and cascading effects occur when destination is busy
Solution Approach 1:
The patent segments the mirror write operation into two independent parts: (1) source filesystem operations that prepare mirror transaction data, and (2) destination filesystem operations that apply the data. By allowing these to proceed independently without blocking, the system eliminates cascading delays while maintaining data protection through transaction logging and replay capabilities.
Solution Approach 2:
The source server performs preliminary actions by preparing mirror transaction data and logging it before the destination is ready to apply it. This preliminary preparation allows the destination to process data at its own pace without blocking the source, resolving the time delay issue while preserving data integrity through the transaction log.
2Loss of time
If mirror data is written to temporary file and replayed later to avoid blocking source, then source delays are prevented, but additional processing and I/O resources are consumed
Solution Approach 1:
The patent extracts the transaction data from the destination filesystem operation and stores it separately as a log file at the source. This extraction allows the source to continue operations without waiting for the destination, eliminating source delays. The destination can then apply transactions from the log at its own pace without requiring complex temporary file storage and replay mechanisms at the destination side.
3Reliability
If full backup strategies are used to preserve all data, then data protection is improved, but backup time and storage requirements increase significantly
Solution Approach 1:
The system performs preliminary actions by continuously logging mirror transactions at the source as they occur, rather than waiting to accumulate large amounts of data before backing up. This continuous logging approach allows for incremental synchronization without requiring time-consuming full backups, as the destination can apply transactions from the log in real-time or near-real-time.
Data Source
AI summary
Methods of operating a computer storage system where an identical copy of a mirror source filesystem is maintained on a mirror destination storage volume managed by a mirror destination server. A modification to the mirror source filesystem is planned and executed, and a corresponding change to the mirror destination filesystem is accomplished by storing data directly to the mirror destination storage volume, bypassing filesystem logic that interprets the data stored on the volume. Systems that implement these methods are also described.