Accelerated RAID Rebuild via Mapped LBA Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As storage capacities increase, the time duration for data rebuild operations in RAID systems also increases, making existing data recovery methods inefficient.
Innovation Solution
The system identifies mapped logical block addresses (LBAs) in a storage system, allowing for accelerated data recovery by reading and writing only the contents of mapped LBAs to a replacement storage device, independent of the file system and partitioning, thereby reducing rebuild time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire address space of a failed storage device is copied to a replacement storage device during rebuild operations, then data recovery completeness is ensured, but the rebuild time duration increases significantly
Solution Approach 1:
The patent extracts and identifies only the subset of logical block addresses (LBAs) that contain valid mapped data from the failed storage device's entire address space. By extracting only the necessary data portions rather than copying the entire address space, the rebuild operation completes faster while still ensuring all valid data is recovered.
Solution Approach 2:
Instead of performing a complete copy of the entire address space (excessive action), the patent performs a partial copy operation that targets only the mapped LBAs that contain valid data. This partial action approach reduces the total data transfer volume and rebuild time while maintaining data recovery completeness.
2Quantity of substance
If storage device capacity increases to provide more storage space, then storage capacity is improved, but the rebuild time duration also increases
Solution Approach 1:
The patent performs preliminary identification of mapped LBAs before the actual rebuild copy operation begins. By pre-identifying which LBAs contain valid mapped data, the system prepares a targeted copy list that prevents unnecessary data transfer during the rebuild process, thus reducing rebuild time even as storage capacities scale up.
Solution Approach 2:
The patent segments the large storage address space into individual logical block addresses and identifies only those segments (LBAs) that contain valid mapped data. This segmentation approach allows the rebuild operation to process only relevant data blocks rather than treating the entire large capacity space as a single unit, thereby reducing rebuild time proportionally.
Data Source
AI summary
One embodiment provides a storage system. The storage system includes storage system control logic to identify at least one target storage device in response to detection of a failed storage device, request a state of a target device logical block address (LBA) from each of the at least one target storage device, and read data associated with a mapped device LBA from each target storage device and write the data to at least one replacement storage device. Another embodiment provides a storage device. The storage device includes device control logic to determine a state of a target device logical block address (LBA) in response to a request; a host interface to provide a reply to the request, the reply including a state indicator related to the state of the target device LBA; a map table including a plurality of device LBAs and respective state indicators; and non-volatile memory (NVM) including data related to at least one mapped LBA.