Stripe Management Using Check-Unit Recombination for EC Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage systems using erasure code technology, achieving a complete stripe is challenging when the number of data units is large, leading to reduced data storage reliability and increased storage costs due to higher storage space utilization.
Innovation Solution
A stripe management method that generates a new check unit using existing check units from multiple stripes, allowing for a conversion from a small erasure code ratio to a large one, facilitating easier stripe completion and reducing redundant data, thereby improving storage space utilization and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a larger quantity of data units is used in EC encoding, then storage space utilization is improved, but it becomes relatively difficult to obtain a complete EC stripe, affecting data storage reliability
Solution Approach 1:
The patent segments the EC stripe completion problem into two phases: first using a smaller data unit quantity (first EC stripe) to easily obtain complete stripes and generate check units, then using a larger data unit quantity (second EC stripe) for final storage with higher space utilization. This segmentation allows each phase to optimize for its specific goal without compromising the other.
Solution Approach 2:
The patent performs preliminary EC encoding with a smaller data unit quantity to generate check units before performing the final EC encoding with a larger data unit quantity. This preliminary action ensures that check units are available to complete the second EC stripe, guaranteeing data storage reliability while achieving high space utilization.
2Quantity of substance
If a larger quantity of data units is used in EC encoding, then storage space utilization is improved, but storage costs increase due to higher redundant data proportion
Solution Approach 1:
The patent changes the erasure code ratio parameter between two phases: using a first erasure code ratio with smaller data unit quantity for preliminary encoding, then switching to a second erasure code ratio with larger data unit quantity for final storage. This parameter change allows optimization of both reliability and space utilization while controlling redundant data through the relationship between the two ratios.
3Reliability
If check units are generated using data units in traditional EC encoding, then fault tolerance is achieved, but computing resources are consumed
Solution Approach 1:
The patent extracts the check unit generation function from the data units in the second EC stripe and performs it separately using check units from the first EC stripe. This extraction avoids redundant computation on data units that will be used for storage, reducing computing resource consumption while maintaining fault tolerance through the generated check units.
Data Source
AI summary
The present disclosure relates to stripe management methods, storage systems, stripe management apparatuses, and storage mediums. In one example method, check units in a plurality of stripes are obtained. A first stripe of the plurality of stripes complies with a first erasure code ratio. A new check unit is generated based on the check units in the plurality of stripes. The new check unit and data units in the plurality of stripes belong to a new stripe, the new stripe complies with a second erasure code ratio, and a quantity of data units corresponding to the first erasure code ratio is less than a quantity of data units corresponding to the second erasure code ratio.