Spare Device Parity Storage for RAID Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAID systems, especially those using SSDs, face challenges in reliability and availability due to limited erasure cycles, read disturb errors, write errors, retention errors, and high latencies caused by internal processes, necessitating enhanced reliability and low latency performance.
Innovation Solution
Implementing a spare device in RAID systems to store additional parity information and metadata, allowing for immediate data reconstruction upon drive failure and reducing latency by utilizing extra parity information stored on the spare device, which is not part of the conventional RAID protocol, thereby improving system reliability and availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional parity-encoding schemes are used in RAID systems, then the system structure is simple, but system reliability is insufficient due to limited erasure cycles and partial failures
Solution Approach 1:
The patent pre-calculates and stores multiple sets of parity information (first parity, second parity, third parity) in advance before any drive failure occurs. When a drive fails, the system can immediately reconstruct data using the pre-stored parity sets without performing complex real-time calculations, thus improving reliability while managing complexity through upfront preparation.
Solution Approach 2:
The patent divides the parity protection into different local sets (first parity set, second parity set, third parity set) stored in different locations or on different spare drives. Each parity set is optimized for specific reconstruction scenarios, allowing the system to handle different failure modes with appropriate local parity information rather than using a single uniform parity scheme.
2Reliability
If data is reconstructed from failed drives using conventional methods, then data can be recovered, but latency is high due to internal processes like garbage collection and wear leveling
Solution Approach 1:
The system pre-calculates multiple parity information sets and stores them on spare drives before any failure occurs. When a drive fails, reconstruction can immediately begin using the pre-stored parity information without waiting for garbage collection or wear leveling to complete, significantly reducing reconstruction latency while maintaining data recovery capability.
Solution Approach 2:
The patent segments the reconstruction process by having multiple spare drives store different parity sets (first parity, second parity, third parity). This allows parallel reconstruction operations where different portions of data can be reconstructed simultaneously from different parity sets, reducing overall reconstruction time and latency.
3Productivity
If SSDs are used in RAID systems, then storage capacity and speed are improved, but reliability decreases due to limited erasure cycles and partial failures
Solution Approach 1:
The patent pre-calculates and stores multiple sets of parity information (first, second, and third parity sets) on spare SSDs before any failure occurs. This allows the system to maintain high storage speed while improving reliability, as the pre-stored parity information can be used for reconstruction without requiring additional erasure cycles on the failed SSD at the moment of failure.
Solution Approach 2:
The patent implements different levels of parity protection (first parity, second parity, third parity) stored in different locations or on different spare SSDs. This allows the system to optimize reliability for different failure scenarios while maintaining the high speed characteristics of SSDs, as each parity set can be independently accessed without affecting the others.
4Reliability
If a spare drive is kept available for immediate reconstruction, then reliability is improved, but the spare drive capacity is wasted during normal operation
Solution Approach 1:
The patent makes the spare drive serve multiple functions: during normal operation, it stores pre-calculated parity information that can be used for future reconstruction; when a failure occurs, it immediately provides the stored parity for data recovery. This multi-functionality allows the spare drive to maintain reliability benefits while its capacity is utilized for storing useful parity data rather than being completely idle.
Solution Approach 2:
The patent changes the state of the spare drive from a completely idle state to an active state where it stores and manages multiple sets of parity information. By dynamically managing the parity sets on the spare drive, the system optimizes the balance between having the drive available for immediate reconstruction and utilizing its capacity for storing valuable parity data that can prevent data loss.
Data Source
AI summary
Efficient use of spare device(s) associated with a group of devices is disclosed, including: receiving a set of data; storing the set of data across a group of devices according to a protocol; determining a set of additional data units associated with the set of data; and storing the set of additional data units in a spare device associated with the group of devices, wherein the spare device is available to be used in the event of a failure of a device included in the group of devices, including by overwriting the set of additional data units as needed.