Distributed LVM Orchestration Without Cluster Storage Bottlenecks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face inefficiencies and high costs due to reliance on centralized storage arrays, which require significant computing resources, are prone to errors, and lack flexibility in managing distributed storage across multiple servers, leading to performance bottlenecks and maintenance challenges.
Innovation Solution
A distributed data storage system utilizing unexploited hardware resources, including Direct Attached Storage (DAS) across multiple servers, with an orchestrator managing a control plane to orchestrate storage stack components, enabling efficient data management, redundancy, and snapshotting without the need for centralized clusters or rescan processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized storage arrays are used to provide data storage and backup capabilities, then data storage management services can be provided, but significant computing resources, infrastructure, and budget capacities are required
Solution Approach 1:
The patent segments the centralized storage array into distributed storage components across multiple servers. Each server manages its own storage locally, eliminating the need for a single centralized array. The storage system is divided into independent units that can operate autonomously, reducing overall infrastructure complexity while maintaining management capabilities.
Solution Approach 2:
The patent implements self-service by allowing each server to manage its own storage resources without requiring centralized array control. Servers perform their own storage management tasks such as allocation, backup, and redundancy independently, eliminating the need for extensive centralized infrastructure and reducing overall system complexity.
2Reliability
If traditional RAID configurations are used to provide data redundancy, then data protection is achieved, but performance penalty on write operations occurs due to parity calculations
Solution Approach 1:
The patent extracts the parity calculation function from the storage array controller and distributes it across multiple server processors. Instead of a single controller handling all parity calculations sequentially, each server's processor handles parity operations for its local storage, parallelizing the computation and eliminating the performance bottleneck associated with centralized RAID controllers.
Solution Approach 2:
The patent dynamically adapts redundancy strategies by allowing flexible configuration of redundancy levels at the server level rather than enforcing fixed RAID levels. Systems can dynamically adjust between different redundancy approaches based on performance needs, enabling optimization between reliability and write performance.
3Ease of operation
If centralized storage arrays are used to provide storage services, then storage management functions are consolidated, but maintenance challenges and errors increase
Solution Approach 1:
The patent segments storage management functions from centralized control to distributed management at each server. This segmentation reduces maintenance complexity by isolating failures to individual servers rather than affecting a centralized array. Each server can be maintained independently, reducing overall maintenance difficulty while preserving management consolidation through standardized interfaces.
Data Source
AI summary
A centralized orchestrator configured to be the only resource managing the distributed LVMs whereby orchestrator is further configured to queue parallel operations on the distributed LVM, making sure only one operation is performed at a time, without a cluster.