Elastic Spare Mechanism for Multi-Failure Data Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data storage systems face challenges in managing multiple failures within a short time frame, as they rely on fixed spare space for data reconstruction, which can be overwhelmed by consecutive failures, leading to increased maintenance costs and urgency in service visits.
Innovation Solution
The implementation of an elastic spare mechanism that dynamically assigns unused space for storing reconstructed information using a layout function, allowing for the creation of spare objects with elastic spare units to handle additional failures beyond the designated spare space, providing partial protection against device failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed spare space is allocated for data reconstruction, then data protection against single failures is ensured, but the system cannot handle multiple consecutive failures
Solution Approach 1:
The patent transforms the static fixed spare space allocation into a dynamic elastic spare mechanism. When failures occur, the system dynamically activates unused space from other nodes to serve as spare capacity. This dynamic reallocation allows the system to adapt to varying failure scenarios, providing protection against multiple consecutive failures while maintaining efficient use of storage resources under normal conditions.
Solution Approach 2:
The patent makes storage space universal by enabling unused space to serve dual purposes: as regular storage capacity during normal operation and as elastic spare capacity when failures occur. This multi-functionality eliminates the need for dedicated fixed spare space, allowing the same physical resources to provide both performance and reliability benefits across different operational states.
2Reliability
If unused space is utilized for storing reconstructed information, then protection against additional failures is provided, but system complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where the storage system automatically manages elastic spares without requiring external intervention. The controller autonomously monitors failure conditions, identifies available unused space, allocates it as elastic spare capacity, and manages the reconstruction process. This self-service approach masks the underlying complexity while providing robust multi-failure protection.
Solution Approach 2:
The system employs feedback mechanisms to continuously monitor the state of storage nodes and failure conditions. Based on this feedback, the controller dynamically adjusts the allocation and activation of elastic spare space. This feedback-driven management simplifies the user experience while enabling complex adaptive behavior for handling multiple failures.
3Productivity
If all unused space is assigned as spare space, then service visit frequency is reduced, but storage efficiency may be compromised
Solution Approach 1:
The patent applies dynamic allocation where unused space remains available for regular storage operations until a failure occurs. Upon failure, the same space is dynamically repurposed as elastic spare capacity for reconstruction. This time-based differentiation allows the system to optimize for storage efficiency during normal operation while providing protection capabilities when needed, reducing service visits without permanent efficiency loss.
Data Source
AI summary
A system utilizing elastic spares includes nodes and devices to store one or more data objects having information. The data object includes data object units each assigned to a storage location on a different node than the other units. The data object includes one or more spare units to store reconstructed information of a failed unit. When one of the data objects has a failed unit and no spare units available to store reconstructed information, a controller of the system assigns an elastic spare unit to an available storage location of one of the nodes. Reconstructed information of the failed unit is stored in the elastic spare unit.