RAID Stripe Mapping Management for Disk Failure Resilience
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional RAID-based storage systems face performance issues due to limited storage space for mapping information elements, leading to degraded access when multiple disks encounter glitches simultaneously, resulting in insufficient space to store new mapping information elements.
Innovation Solution
The method involves allocating storage space for mapping information elements and assigning sequence numbers to manage storage efficiently, allowing for the override of existing elements when space is insufficient, ensuring continuous access to RAID stripes even during disk failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If limited storage space is used for mapping information elements, then device complexity is reduced, but reliability deteriorates when multiple disks fail simultaneously
Solution Approach 1:
The patent implements dynamic management of mapping information elements by introducing sequence numbers and override mechanisms. When disk failures occur, the system dynamically updates mapping information by overwriting older entries with newer ones that reflect current disk states, allowing the limited storage space to adapt to changing conditions rather than maintaining a static structure
Solution Approach 2:
The patent changes the parameter of mapping information management by introducing sequence numbers as a new attribute. This allows the system to track the temporal order of mapping information elements and use this parameter to determine which entries to override during disk failures, transforming the storage space from a simple container to a temporally-aware structure
2Reliability
If more storage space is allocated for mapping information elements, then reliability improves during disk failures, but device complexity increases
Solution Approach 1:
The patent segments the storage space for mapping information elements into discrete entries, each with a sequence number. This segmentation allows the system to manage limited space efficiently by overwriting only specific entries when needed, rather than requiring a completely separate backup storage structure for every possible failure scenario
3Ease of operation
If mapping information elements are stored without sequence numbers, then ease of operation is improved, but productivity deteriorates when space is insufficient
Solution Approach 1:
The patent applies preliminary action by assigning sequence numbers to mapping information elements at the time of their creation. This pre-labeling allows the system to quickly identify which entries to override when disk failures occur, eliminating the need for complex search operations and maintaining fast I/O performance even under failure conditions
Data Source
AI summary
A technique involves: in response to a stripe being created in a Redundant Array of Independent Disks (RAID), allocating to the stripe a storage space for storing a predetermined number of mapping information elements; creating one or more mapping information elements for the stripe; storing the one or more mapping information elements in the storage space; allocating respective storage sequence numbers to the one or more mapping information elements stored in the storage space; and in response to receiving an I/O request for the stripe indicating one of the respective storage sequence numbers, accessing the stripe based on the mapping information element in the storage space that is corresponding to the storage sequence number indicated by the I/O request. Such a technique can facilitate when disks encounter “glitch issues” simultaneously under extreme situations and effectively improve I/O read and write performance of a storage system.