Dynamic Overprovisioned Space Rebuild for Storage Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage systems face challenges in efficiently rebuilding a failed storage device, particularly in determining where to store recovered data, leading to degraded system performance and increased costs due to the need for hot spares or complex thin-provisioned environments.
Innovation Solution
The method involves designating a system-level overprovisioned (OP) space in the storage system, where data from a failed device is reconstructed and stored in dynamically selected locations based on performance and reliability criteria, reducing the time in degraded mode and implementation complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a hot spare storage device is reserved to store recovered data, then the time in degraded mode is reduced, but the hardware cost increases
Solution Approach 1:
The patent makes existing storage devices in the array serve multiple functions: they store both their own data and can be dynamically used to store recovered data from failed devices. This eliminates the need for dedicated hot spare devices while maintaining the ability to quickly rebuild failed storage units.
Solution Approach 2:
The storage system uses its own existing resources (available space in other storage devices) to rebuild failed devices, rather than relying on external dedicated spares. The system self-manages the rebuild process by dynamically selecting and using appropriate target locations within the array.
2Quantity of substance
If thin-provisioned environment is used to dynamically decrease capacity for rebuild, then hardware cost is reduced, but implementation complexity and handling complications increase
Solution Approach 1:
The storage system automatically manages the dynamic allocation and selection of target locations for recovered data without requiring complex thin-provisioning infrastructure. The controller autonomously identifies suitable storage locations based on available space and system state, simplifying implementation while maintaining dynamic capacity management.
3Productivity
If data is stored in dynamically selected locations in overprovisioned space, then operational cost is reduced and degraded mode time is minimized, but system complexity increases
Solution Approach 1:
The patent implements dynamic selection of target storage locations based on real-time system state, including which devices are available and have sufficient space. This dynamic approach optimizes rebuild operations by adaptively choosing the best target locations, improving operational efficiency while using manageable complexity through algorithmic selection based on available information.
Solution Approach 2:
The system changes the parameter of target location selection from static (fixed hot spare) to dynamic (based on real-time device availability and space). This allows the system to adapt to different failure scenarios and optimize rebuild performance by selecting from multiple potential target locations based on current system parameters.
Data Source
AI summary
Methods and systems for rebuilding a failed storage device in a storage system using a plurality of dynamically selected locations in a system-level overprovisioned (OP) space. For example, a method including designating a portion of a usable space in a storage system as a system-level OP space; in response to a detection of a failed storage device in the storage system, reconstructing data of the failed storage device based on data read from a plurality of functioning storage devices in the storage system; and storing the reconstructed data of the failed storage device in a plurality of dynamically selected locations in the system-level OP space of the plurality of functioning storage devices.