Storage Virtualization Metadata Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage virtualization environments, the loss of a vendor provider (VP) or its metadata database can disrupt service continuity, as virtual machines lack the necessary metadata to locate stored data, leading to operational issues.
Innovation Solution
Implementing recovery support techniques by backing up VP metadata, including vvol, container, and SCP metadata, and storing it within the storage system, allowing for the reconstruction of metadata in case of VP failure, using a replacement VP and multi-threaded approaches to maintain system performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If VP metadata is stored only in the VP's database, then the system structure is simple, but service continuity is disrupted when VP fails
Solution Approach 1:
The patent implements preliminary backup of VP metadata to the storage system before failure occurs. The metadata is periodically or incrementally synchronized from the VP's database to the storage system, ensuring recovery capability is established in advance. This resolves the contradiction by preparing recovery resources beforehand without requiring complex real-time redundancy structures.
Solution Approach 2:
The storage system acts as an intermediary backup target for VP metadata. Instead of directly replicating between multiple VPs or using complex distributed storage, the storage system serves as a centralized intermediary that receives and preserves metadata copies. This simplifies the overall structure while ensuring service continuity through the intermediary backup mechanism.
2Reliability
If full metadata backup is performed, then recovery completeness is improved, but recovery time increases
Solution Approach 1:
Metadata is backed up incrementally or periodically before failure occurs, so that much of the backup work is already completed in advance. When recovery is needed, only the most recent changes need to be transferred, significantly reducing actual recovery time while maintaining complete metadata recovery capability.
Solution Approach 2:
The system performs partial metadata backups at intervals rather than complete backups continuously. This partial action approach maintains recovery completeness by capturing essential metadata changes over time, while reducing the time and resources required for each backup operation and subsequent recovery.
3Reliability
If metadata is stored in distributed locations, then system availability is improved, but metadata consistency becomes difficult to maintain
Solution Approach 1:
The storage system serves as a centralized intermediary that receives and stores metadata backups from the VP. This centralized intermediary approach maintains metadata consistency by providing a single authoritative backup location, while still improving system availability through the backup mechanism. The intermediary simplifies consistency management compared to fully distributed storage.
Solution Approach 2:
The system implements feedback mechanisms to track and verify metadata backup status. The VP monitors whether metadata has been successfully backed up to the storage system and can trigger additional backups or alerts if consistency checks fail. This feedback loop maintains metadata consistency across the distributed VP-storage system without requiring complex distributed consensus protocols.
Data Source
AI summary
Recovery support techniques for storage virtualization environments are described. In one embodiment, for example, a method may be performed that comprises defining, by processing circuitry, a storage container comprising one or more logical storage volumes of a logical storage array of a storage system, associating the storage container with a virtual volume (vvol) datastore, identifying metadata for a vvol of the vvol datastore, and writing the metadata for the vvol to the storage system. Other embodiments are described and claimed.