Dynamic Storage Mapping via Hash Function for Rebuilding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional RAID technologies are limited by the writing input/output bandwidth of hot spare disk drives, making it difficult to reduce rebuilding time and leading to increased memory consumption and software complexity due to the need for storing mapping tables.
Innovation Solution
The method involves dynamically determining a mapping between logical storage extents and storage devices, distributing logical blocks based on this mapping, and using a hash function to reduce memory consumption by not storing the mapping table, allowing only changed mappings to be stored and released when a faulty device is replaced.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional RAID technology uses a hot spare physical disk drive for data rebuilding, then data redundancy and reliability are maintained, but the rebuilding time is limited by the writing input/output bandwidth of the hot spare disk drive
Solution Approach 1:
The patent segments the rebuilding process by distributing data blocks across multiple storage devices in a storage pool rather than relying on a single hot spare disk. The logical storage extent is divided and reconstructed in parallel across multiple devices, breaking the single-bottleneck constraint and enabling concurrent data reconstruction operations.
Solution Approach 2:
The patent transforms the traditional single hot spare disk into a multi-functional storage pool where any available storage device can participate in data reconstruction. The system dynamically selects multiple storage devices from the pool to perform rebuilding operations, making the storage infrastructure universally applicable for both primary storage and reconstruction purposes.
2Ease of operation
If mapping tables are stored to manage the distribution of logical storage extents to storage devices, then data distribution management is achieved, but memory consumption increases
Solution Approach 1:
The patent extracts the mapping table storage requirement from the system by using a hash function to dynamically determine the distribution of logical storage extents to storage devices. Instead of storing the entire mapping table in memory, the system computes the mapping on-demand through hash operations, eliminating the need for large memory allocations while maintaining distribution management capabilities.
Solution Approach 2:
The hash function serves itself to generate mapping information without requiring external storage of mapping tables. The system uses the hash function to self-determine the distribution of data blocks across storage devices, and when a device fails, the same hash function automatically redistributes the data without requiring pre-stored mapping information.
Data Source
AI summary
Embodiments of the present disclosure relate to a method, a device and a computer readable medium for managing a storage system. The method can determine a mapping between a logical storage extent and a plurality of storage devices in a storage pool without storing a table of the mapping. Further, the method can record a new mapping after a storage devices fails and restore the original mapping after the failed storage device is replaced.