RAID Track Aggregation for Backup Volume Contiguity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data mirroring technologies in RAID storage systems face inefficiencies when copying non-contiguous tracks to backup volumes, leading to increased drive operations and processing time due to the RAID penalty, especially when writing less than a full stride.
Innovation Solution
The implementation of a container data structure that aggregates logically random tracks into physically contiguous groups, allowing for efficient writing of tracks as a full stride with updated parity, reducing the number of drive operations and processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If tracks are copied individually to backup volume, then data redundancy is maintained, but the number of drive operations increases and processing time increases
Solution Approach 1:
Multiple individual track copy operations are merged into a single aggregated stride copy operation. The system collects multiple tracks that need to be copied and combines them into one RAID stride unit, which is then copied in a single operation to the backup volume. This merging reduces the total number of drive operations while maintaining data redundancy, as the aggregated stride contains all the necessary track data.
Solution Approach 2:
The system performs preliminary aggregation of tracks into a container data structure before the actual copy operation to backup volume. By pre-grouping tracks that will be copied together into a single stride unit, the system prepares the data in advance, reducing the number of subsequent drive operations required during the backup process.
2Ease of operation
If tracks are stored non-contiguously on backup volume, then logical track mapping is simplified, but RAID penalty increases and processing time increases
Solution Approach 1:
The system merges multiple logical tracks into a single physical RAID stride unit that is stored contiguously on the backup volume. By combining tracks into an aggregated stride container, the system achieves physically contiguous storage that eliminates the RAID penalty associated with non-contiguous writes, while the container data structure maintains the logical mapping relationships.
Solution Approach 2:
The system introduces an intermediate aggregation dimension by grouping tracks into RAID stride units. Instead of directly mapping individual logical tracks to physical locations, the system creates a intermediate layer where tracks are aggregated into strides, which are then stored contiguously. This dimensional transformation resolves the conflict between logical simplicity and physical efficiency.
3Productivity
If aggregation of tracks is performed, then number of drive operations is reduced, but container data structure complexity increases
Solution Approach 1:
The container data structure serves multiple functions simultaneously: it aggregates multiple tracks into a single unit, maintains the mapping between logical tracks and physical stride locations, and manages the aggregation process itself. By making the container multi-functional, the system reduces drive operations without requiring separate complex structures for each function, thereby limiting the increase in overall complexity.
Data Source
AI summary
A plurality of tracks that are to be copied to a backup volume are aggregated in a container data structure. The plurality of tracks are stored physically contiguously in a single Redundant Array of Independent Disks (RAID) stride. Mapping metadata is updated in the backup volume to indicate how logical tracks of the backup volume correspond to physical tracks stored in the RAID stride.