Mapped RAID Proactive Copy Parallel Extent Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data storage appliances using RAID technology face performance limitations during the rebuild process after a drive failure, which can lead to permanent data loss if another drive fails before the rebuild is completed, and are bottlenecked by write throughput and metadata overhead during proactive copy operations.
Innovation Solution
The implementation of specialized mapped RAID techniques, where data is distributed across RAID extents on different physical storage drives, allowing for parallel copying of disk extents and reducing metadata overhead, thereby utilizing the faster read bandwidth of the failing drive and minimizing I/O operations during the proactive copy process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is proactively copied to a hot spare drive using conventional RAID, then data protection is improved, but write throughput becomes a bottleneck limiting rebuild speed
Solution Approach 1:
The patent segments the RAID volume into multiple independent disk extents, each of which can be proactively copied to different destination drives in parallel. This segmentation allows the system to bypass the single hot spare bottleneck by distributing copy operations across multiple destination drives, thereby improving rebuild speed while maintaining data protection.
Solution Approach 2:
The patent transitions from a single-destination hot spare model to a multi-destination parallel copy model. By introducing multiple destination drives and copying disk extents to different destinations simultaneously, the system adds a dimensional aspect to the proactive copy process, eliminating the write throughput bottleneck of a single hot spare drive.
2Reliability
If conventional RAID proactive copy is used, then data protection is improved, but metadata overhead increases significantly
Solution Approach 1:
The patent extracts the metadata management burden from the proactive copy process by using Mapped RAID's inherent extent mapping capabilities. Instead of creating new metadata to track proactive copy operations, the system leverages existing Mapped RAID metadata structures, thereby reducing metadata overhead while maintaining data protection.
Solution Approach 2:
The patent makes the Mapped RAID metadata structure serve multiple functions: it continues to manage normal RAID operations while also tracking proactive copy operations. This multi-functionality eliminates the need for separate metadata structures for proactive copying, reducing overall metadata overhead.
3Ease of operation
If I/O operations continue on the failing drive during PACO, then data availability is maintained, but overhead increases due to metadata management
Solution Approach 1:
The patent performs preliminary mapping updates before the proactive copy completes. By pre-assigning disk extents to hot spares and updating mappings in advance, the system reduces the complexity of managing I/O operations during the copy process, as the mapping structure already reflects the intended data location.
Solution Approach 2:
The Mapped RAID metadata structure automatically manages the complexity of tracking proactive copy operations and I/O redirection. The self-updating metadata system handles the overhead of managing data availability during PACO without requiring additional complex management mechanisms.
4Productivity
If parallel copying of disk extents is implemented, then rebuild speed is improved, but device complexity increases
Solution Approach 1:
The patent segments the RAID volume into multiple independent disk extents that can be copied in parallel to different destination drives. This segmentation enables speedup proportional to the number of parallel copy operations, while the modular nature of extent-based segmentation keeps system complexity manageable through the existing Mapped RAID framework.
Solution Approach 2:
The patent introduces parallelism as a new dimension to the proactive copy process by copying multiple disk extents simultaneously to different destinations. This dimensional expansion of the copy operation increases rebuild speed while the underlying Mapped RAID infrastructure manages the complexity of coordinating these parallel operations.
Data Source
AI summary
A method performed in a Mapped RAID system includes: (a) in response to a source disk (SD) reaching EOL status: (1) copying a first source disk extent (DE) (belonging to a first RAID extent) from the SD to a first spare DE on a first destination disk and (2) copying a second source DE from the SD to a second spare DE (belonging to a second RAID extent) on a second destination disk, (b) in response to completely copying the first source DE from the SD to the first spare DE, mapping the first spare DE to belong to the first RAID extent in place of the first source DE, and (c) in response to completely copying the second source DE from the SD to the second spare DE, mapping the second spare DE to belong to the second RAID extent in place of the second source DE.