Storage Metadata Recovery Using Remote Backup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional metadata recovery methods in storage systems require reserving metadata snapshots locally and remotely, leading to high storage overhead and capacity issues, especially in multi-site environments, as they occupy excessive space and hinder efficient resource utilization.
Innovation Solution
A method that utilizes remote backup metadata for recovery without reserving metadata snapshots locally, allowing only the latest metadata to be stored on each device, using journals and a B+ tree structure to ensure data integrity and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If metadata snapshots are reserved locally and remotely for recovery, then data reliability is improved, but storage overhead increases
Solution Approach 1:
The patent extracts the metadata snapshot storage function from the local device and relocates it to a remote device. Only the latest metadata is kept locally, while historical metadata snapshots are stored remotely. This separation reduces local storage overhead while maintaining recovery capability through remote access to historical snapshots.
Solution Approach 2:
The patent introduces a remote storage dimension for metadata snapshots, moving from a single-location (local) storage model to a distributed model across multiple dimensions (local latest metadata + remote historical snapshots). This dimensional expansion allows recovery functionality without proportionally increasing local storage requirements.
2Reliability
If multiple metadata snapshots are retained for recovery in multi-site environments, then recovery reliability is improved, but storage capacity consumption increases
Solution Approach 1:
The patent applies different quality levels to different locations: the local device maintains only the latest metadata with high accessibility, while the remote device stores historical snapshots with lower accessibility but sufficient for recovery. This local quality differentiation optimizes the balance between recovery reliability and storage capacity consumption.
3Speed
If metadata snapshots are stored locally for fast recovery, then recovery speed is improved, but storage overhead increases
Solution Approach 1:
The patent prepares the latest metadata locally in advance for immediate recovery access, while historical snapshots are pre-stored remotely. This preliminary action ensures that the most critical recovery scenario (latest metadata restoration) can proceed at high speed from local storage, while less frequent historical recoveries use remote storage.
Data Source
AI summary
Embodiments of the present disclosure relate to a method, device and computer program product for recovering metadata. The method comprises detecting whether first metadata corresponding to a first journal at a first device is corrupted, and obtaining second metadata corresponding to a second journal from a second device in the case that the first metadata is corrupted, wherein the second device is remote from the first device. The method further includes recovering the first metadata based on the second metadata, the first journal and the second journal. Embodiments of the present disclosure utilize remote backup metadata for recovery in the case that local metadata (which for example includes index information) is corrupted without reserving metadata snapshot locally, thereby saving the storage space.