Block Storage Virtualization Modules for Coherent Sharing and High Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer systems face poor capacity utilization, duplication of storage management efforts, and single points of failure due to reliance on either direct-attached storage (DAS) or Storage Area Networks (SAN), which do not effectively enable coherent sharing of logical disks and redundancy across multiple systems.
Innovation Solution
Pairing discrete block storage virtualization modules on each computer system with a shared block storage virtualization module to maintain local and global logical volume coherency, providing high availability, data protection, and centralized storage management using a shared node, allowing for efficient replication and access to disk resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple computer systems use direct-attached storage (DAS) configurations, then each system has independent storage capacity and isolation, but capacity utilization is poor and storage management effort is duplicated
Solution Approach 1:
The patent combines multiple DAS systems with a SAN system by implementing storage virtualization modules on each DAS system that can present storage resources to multiple systems. This merging allows storage resources to be shared across systems while maintaining the independence of individual DAS units, thereby improving capacity utilization without sacrificing system independence.
Solution Approach 2:
The storage virtualization module implements multi-functionality by enabling storage resources to serve both local system needs and shared network needs. The same physical storage can be accessed independently by its host system while simultaneously being available to other systems through the SAN interface, eliminating the need for separate storage allocations.
2Productivity
If multiple computer systems connect to a common Storage Area Network (SAN), then centralized storage management and resource sharing are achieved, but a single point of failure is created
Solution Approach 1:
The patent segments the storage system into multiple independent components - individual DAS systems, each with its own storage virtualization module, connected to the SAN. This segmentation ensures that a failure in one DAS system or SAN component does not affect others, as each system maintains its own local storage resources that can operate independently.
Solution Approach 2:
The system implements prior cushioning by maintaining local storage resources on each DAS system that can serve as backup or alternative access paths. If the SAN fails, each system can continue to access its local storage resources, providing a cushion against SAN failures and eliminating the single point of failure.
3Reliability
If external storage systems are mirrored between two or more physically-separated subsystems, then redundancy and high availability are achieved, but device complexity and infrastructure cost increase
Solution Approach 1:
The storage virtualization module implements self-service by automatically managing the mirroring and redundancy operations. The system autonomously handles data replication, failover, and recovery without requiring complex manual configuration or intervention, thereby achieving redundancy while reducing the operational complexity.
Solution Approach 2:
The storage virtualization module acts as an intermediary that simplifies the complexity of redundancy management. It provides a unified interface for managing mirrored storage resources across multiple systems, abstracting away the complex infrastructure details from users and administrators.
4Speed
If local direct-attached storage is maintained on each server, then data access speed is fast, but storage management effort is duplicated across systems
Solution Approach 1:
The storage virtualization module serves as an intermediary that manages storage resources centrally while maintaining fast local access. It coordinates storage operations across multiple systems, eliminating duplicated management efforts, while preserving the speed benefits of local DAS through cached or frequently accessed data.
Data Source
AI summary
Collaboration between discrete systems and a shared system to consolidate shared storage-related services. In one example, shared, consolidated storage-related services and high-availability are provided by pairing each of a plurality of discrete block storage virtualization modules residing on each discrete computer system to a shared block storage virtualization module residing on a shared computer system, and maintaining logical volume coherency locally by each of the plurality of discrete block storage virtualization modules, and globally by the shared block storage virtualization module. Additionally, allocation of a set of block storage virtualization functions to be performed by at least one of the plurality of discrete block storage virtualization modules or the shared block storage virtualization module can be made according to the pairing.