Storage Pool Redistribution via Logical Pair Binding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face challenges in efficiently managing and redistributing physical storage units across multiple storage devices while maintaining data resilience and availability, especially during hardware upgrades or failures.
Innovation Solution
The method involves redistributing existing physical storage units across a pool of storage devices by defining logical pairs, binding them to resiliency sets, and performing restriping to ensure each unit spans only a single resiliency set, allowing for even distribution and redundancy across multiple devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If physical storage units are redistributed across multiple storage devices to improve data resilience, then data availability is improved, but system complexity increases due to the need for logical pairs, resiliency sets, and restriping operations
Solution Approach 1:
The storage system is segmented into logical pairs of storage devices, where each pair is bound to a resiliency set. This segmentation allows the system to manage data distribution in controlled units, improving reliability through structured redundancy while keeping the complexity manageable through modular organization.
Solution Approach 2:
Logical pairs serve as intermediaries between physical storage devices and resiliency sets. By introducing this intermediate layer, the system can manage the complexity of data redistribution and restriping operations without directly managing individual devices, thus improving data availability while containing system complexity.
2Reliability
If restriping is performed to ensure each physical storage unit spans only a single resiliency set, then data resilience is improved, but processing time increases during the redistribution operation
Solution Approach 1:
Storage devices are organized into logical pairs and bound to resiliency sets before data redistribution begins. This preliminary organization establishes the target structure for restriping operations, allowing the system to efficiently move data to the correct locations without complex real-time decision-making, thus improving data resilience while minimizing processing time.
3Reliability
If multiple logical pairs are created and bound to different resiliency sets, then data redundancy is improved, but the difficulty of managing and redistributing storage units increases
Solution Approach 1:
The logical pair structure serves multiple functions simultaneously: it provides the basis for data striping, defines resiliency set boundaries, and enables efficient restriping operations. This multi-functionality improves data redundancy while simplifying management by eliminating the need for separate mechanisms to handle these different aspects.
Data Source
AI summary
Techniques for storage management may include: adding first storage devices to a pool, wherein prior to adding the first storage devices to the pool, the pool includes second storage devices with existing physical storage units (PUs) distributed across the second storage devices; and in response to adding the first storage devices to the pool, performing processing to evenly distribute the existing PUs among the pool of storage devices, said processing including: defining logical pairs of storage devices in accordance with rules; binding each of the logical pairs to a resiliency set; determining that a first PU of the existing PUs includes disk slices that span across more than a single resiliency set; andin response to determining that the first PU includes disk slices that span across more than a single resiliency set, performing second processing to restripe the first PU in accordance with a target layout.