Virtual Disk RAID Storage with Protection Pool
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional RAID arrays are confined to a single storage node and a set of physical disks, limiting their scalability and flexibility in allocating and accessing disks, which restricts their ability to efficiently manage data across multiple nodes and disks.
Innovation Solution
The implementation of a RAID system that uses virtual disks instead of physical disks, allowing data to be spread across multiple nodes in a node cluster, with each virtual disk comprising physical extents (PEs) that can be allocated and managed by a local disk manager, enabling multiple RAID arrays to share physical disks and tolerate failures by distributing PEs across different disks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional RAID arrays are confined to a single storage node and set of physical disks, then device simplicity is maintained, but scalability and flexibility are limited
Solution Approach 1:
The patent divides the RAID array into virtual disks that are logically separated from physical disks. Each virtual disk can be independently allocated to different RAID arrays, and physical disks can be shared across multiple virtual disks. This segmentation enables flexible resource allocation and scaling without requiring complete reconfiguration of the storage system.
Solution Approach 2:
The patent introduces a virtualization layer between the physical storage infrastructure and RAID arrays, creating an additional dimension of abstraction. This virtual disk layer allows multiple RAID arrays to access the same physical disks simultaneously through virtualized interfaces, enabling scalable growth without linear increases in physical infrastructure complexity.
2Productivity
If physical disks are shared across multiple RAID arrays, then resource utilization efficiency is improved, but failure tolerance requirements increase
Solution Approach 1:
By segmenting physical disks into multiple virtual disks through the virtualization layer, the patent enables isolated failure domains. When a physical disk fails, only the specific virtual disk allocations on that disk are affected, while other virtual disks on the same physical disk remain operational. This segmentation limits failure propagation and simplifies recovery procedures.
Solution Approach 2:
The virtual disk layer acts as an intermediary between physical disks and RAID arrays, providing abstraction and isolation. This intermediary layer manages failover and data redistribution transparently, allowing the system to tolerate disk failures while maintaining resource sharing. The virtualization layer handles the complexity of failure management, enabling high resource utilization without proportionally increasing failure tolerance requirements.
3Reliability
If data is spread across multiple nodes in a cluster, then availability and reliability are enhanced, but system complexity and coordination overhead increase
Solution Approach 1:
The virtual disk layer provides a universal interface that abstracts the underlying distributed storage infrastructure. Multiple RAID arrays across different nodes can access the same virtual disks through standardized protocols, enabling data distribution and replication without requiring node-specific configurations. This universality simplifies cross-node coordination while maintaining high availability.
4Adaptability or versatility
If virtual disks are used instead of physical disks, then allocation flexibility is improved, but management complexity increases
Solution Approach 1:
The virtual disk management system provides self-service capabilities through automated allocation, mapping, and deallocation of virtual disk segments. The system can dynamically assign physical extents to virtual disks based on available capacity and RAID array requirements without manual intervention. This automation reduces the operational burden of managing virtualized storage while maintaining high allocation flexibility.
Data Source
AI summary
A storage node can include one or more processors and one or more storage disks, where the one or more storage disks include one or more local physical extents (PEs) that are local to the storage node. The storage node can include a protection pool driver executed by the one or more processors to run in a kernel space of the storage node, where the protection pool driver includes a local disk manager (LDM) and an array group module (AGRP). The LDM can be configured to manage the one or more local PEs at the one or more storage disks. The AGRP can include a number of storage arrays, where each of the storage arrays includes one or more virtual disks, where each of the one or more virtual disks is associated to at least a local PE or an external PE external to the storage node.