Peer Storage Data Scrub via Direct Device Communication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-device storage systems face bottlenecks in data scrubbing due to centralized control plane management, leading to underutilization of compute resources and reduced scalability, especially in flash memory-based systems prone to data corruption.
Innovation Solution
Implementing peer-to-peer communication among storage devices to offload data scrub operations, allowing storage devices to communicate directly and manage data scrubbing processes independently of a storage control plane, utilizing a data scrub module and peer communication channel to identify and correct errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data scrub operations are managed by the storage control plane, then data integrity can be maintained, but processing bottlenecks and reduced scalability occur
Solution Approach 1:
The patent extracts the data scrub operation from the centralized storage control plane and relocates it to individual storage devices. Each storage device executes data scrub locally on its own data segments, eliminating the processing bottleneck at the control plane while maintaining data integrity through distributed error detection and correction operations.
Solution Approach 2:
Storage devices perform self-service data scrub operations independently without requiring continuous control plane intervention. Each device monitors its own data segments, executes scrub operations when needed, and manages its own error correction, thereby improving scalability and throughput while maintaining reliability.
2Ease of operation
If storage devices communicate through the storage control plane, then centralized coordination is achieved, but communication bottlenecks and reduced scalability occur
Solution Approach 1:
The patent segments the centralized communication architecture into distributed peer-to-peer communication channels. Storage devices establish direct communication paths with each other for coordinating data scrub operations, eliminating the single-point communication bottleneck at the control plane and reducing communication overhead as the system scales.
Solution Approach 2:
The patent introduces a new communication dimension by enabling direct peer-to-peer interactions between storage devices alongside the existing control plane communication path. This multi-dimensional communication approach allows devices to coordinate scrub operations directly without funneling all communication through the control plane, reducing bottlenecks and improving scalability.
3Adaptability or versatility
If compute resources are centralized at the storage control plane, then system-wide coordination is possible, but resource underutilization at storage devices occurs
Solution Approach 1:
The patent applies local quality by enabling storage devices to execute data scrub operations locally using their own compute resources. Each device utilizes its local processor and memory to perform error detection and correction on its data segments, thereby improving overall compute resource utilization across the storage array while maintaining system-wide coordination capabilities through peer communication.
Data Source
AI summary
Example peer storage systems, storage devices, and methods provide data scrub using a peer communication channel. Peer storage devices establish peer communication channels that communicate data among the peer storage devices. A storage device may identify data segments from their storage medium for a data scrub process. A peer storage device may be identified that contains corresponding data segments to the data segment being scrubbed. A corresponding lock command may be sent over the peer communication channel to lock the corresponding data segments during the data scrub process. A data scrub error report may be generated from the data scrub process. If an error is discovered during the data scrub process the storage device may use the peer communication channel to retrieve recovery data from peer storage devices to rebuild the data segment with the error.