RAID Controller Fast Recovery via Secondary Strip Mirroring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current RAID systems face performance degradation and increased rebuild times during drive failures, leading to potential data loss, especially in scenarios like DRAID-6 where data protection is insufficient upon multiple drive failures.
Innovation Solution
The method involves identifying unused blocks in a RAID array to create redundant copies of high-value host writes, mirroring primary strips across different storage volumes, and maintaining primary-secondary strip mappings to ensure data availability and reduce rebuild times by allowing access to secondary copies in case of drive failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If RAID systems use traditional rebuild methods after drive failure, then data protection is maintained through parity restoration, but rebuild times are prolonged and system performance degrades
Solution Approach 1:
The patent creates secondary copies of high-value data blocks in advance and stores them in unused blocks within the RAID array before failures occur. When a drive fails, these pre-positioned secondary copies can be immediately accessed, eliminating the need for time-consuming rebuild operations from parity data.
2Productivity
If RAID systems distribute rebuild capacity across all drives as in DRAID, then rebuild speed improves, but data protection is insufficient when multiple drives fail concurrently
Solution Approach 1:
The patent creates secondary copies of critical data blocks and stores them in unused blocks within the array. These secondary copies provide an additional layer of redundancy that protects against multiple concurrent drive failures, while the distributed rebuild capacity of DRAID maintains fast rebuild speeds.
3Quantity of substance
If RAID systems allocate all storage blocks for primary data, then storage capacity is maximized, but no redundant copies exist for fast recovery after failures
Solution Approach 1:
The patent makes unused blocks within the existing RAID array serve a dual purpose: they remain available for future data allocation while simultaneously functioning as storage locations for secondary copies of high-value data blocks, enabling fast recovery without sacrificing storage capacity.
Data Source
AI summary
The disclosure includes a computer-implemented method for providing fast data access after a drive failure, a computer program product, and a RAID controller. One embodiment may comprise identifying a RAID array, the RAID array comprising a plurality of storage volumes, identifying an unused block of a provisioned volume in the RAID array, and copying a redundant copy of high value host writes to the unused block. The copying may comprise, for primary strips in the RAID array, creating one or more secondary strips mirroring the primary strips such that each of pair of primary-secondary strips reside on different storage volumes from each other.