Modular Storage Resiliency Groups to Reduce Redundant Writes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies in managing data resilience and storage operations, particularly in modular storage systems, leading to suboptimal performance and reliability due to redundant write operations and lack of centralized control over storage devices.
Innovation Solution
A method for optimizing resiliency group formation in modular storage systems by using a storage controller to identify available storage devices and determine a minimum resiliency level, forming resiliency groups that manage data storage operations without relying on individual storage device controllers, thereby reducing redundant write operations and enhancing system reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If individual storage device controllers are used to manage data storage operations, then each device can operate independently, but redundant write operations increase and system reliability decreases
Solution Approach 1:
The system segments storage devices into resiliency groups managed by a centralized controller, where each group contains a subset of storage devices. This segmentation allows independent management of failure domains while maintaining centralized coordination for reliability, resolving the contradiction between independent operation and system reliability.
Solution Approach 2:
A centralized storage controller acts as an intermediary between storage devices and host systems, coordinating write operations across resiliency groups. The controller manages data distribution and redundancy allocation, eliminating redundant writes while maintaining independent device operation through the intermediary layer.
2Adaptability or versatility
If individual storage device controllers manage data storage operations, then device autonomy is maintained, but redundant write operations increase
Solution Approach 1:
The system dynamically allocates data and redundancy across resiliency groups based on current device availability and failure domains. The centralized controller adjusts write operations in real-time, directing writes only to necessary devices within resiliency groups, thereby maintaining device autonomy while eliminating redundant writes to failed or unavailable devices.
Solution Approach 2:
The system changes the parameter of write operation targeting by implementing resiliency group membership and failure domain awareness. Writes are directed to specific resiliency groups rather than all devices, changing the behavior from universal writes to targeted writes, thus reducing redundant operations while preserving device autonomy.
3Reliability
If centralized control is implemented over storage devices, then redundant write operations are reduced, but system complexity increases
Solution Approach 1:
The centralized controller is segmented into multiple resiliency group managers, each responsible for a specific group of storage devices. This segmentation distributes the control complexity across multiple manageable units rather than requiring a single monolithic controller, reducing the perceived system complexity while maintaining centralized coordination for reliability.
Solution Approach 2:
The system introduces a new dimension of organization by arranging storage devices in a two-dimensional structure: resiliency groups as rows and failure domains as columns. This dimensional organization simplifies the control logic by providing clear rules for data placement and redundancy allocation, reducing the complexity of centralized control while enhancing reliability.
Data Source
AI summary
A storage configuration of a storage system is received by a storage controller and a set of storage devices of a plurality of storage nodes of the storage system that are available within the storage configuration are identified. A minimum resiliency for data stored in the storage system is determined and at least one resiliency group within the storage system based on the available storage devices in the storage configuration of the storage system and the minimum resiliency for data stored in the storage system is generated.