Virtual Storage System Sizing via Component Replacement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face challenges in efficiently managing data storage and retrieval across multiple storage nodes, particularly in ensuring data availability and reliability in the event of node failures.
Innovation Solution
The proposed solution involves a distributed storage system architecture that utilizes multiple storage nodes with non-volatile solid-state storage units, implementing erasure coding and redundant metadata distribution to ensure data availability and reliability. Additionally, the system employs a cloud-based storage approach with local caching to optimize performance and durability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is stored on a single storage node, then storage simplicity is maintained, but data availability and reliability deteriorate when node failures occur
Solution Approach 1:
The patent divides data into multiple segments or chunks and distributes them across multiple storage nodes. This segmentation allows the system to maintain data availability even when individual nodes fail, as data can be retrieved from remaining nodes. The data is split into k segments where any k segments can reconstruct the original data, enabling fault tolerance without requiring complete node redundancy.
Solution Approach 2:
The patent implements hierarchical storage architecture where data is organized in nested structures across multiple levels - from individual storage nodes to storage pools and ultimately to the distributed storage system. This nested structure allows data to be accessible at multiple levels of abstraction, improving reliability while managing complexity through layered organization.
2Reliability
If data is replicated across multiple storage nodes, then data availability is improved, but storage efficiency deteriorates due to redundant data copies
Solution Approach 1:
The patent changes the fundamental parameter of data representation by using erasure coding instead of simple replication. Instead of storing complete duplicate copies of data, the system transforms data into encoded segments where mathematical relationships allow reconstruction of original data from any sufficient subset of segments. This parameter change enables achieving the same reliability with significantly reduced storage overhead.
Solution Approach 2:
The patent uses intelligent copying through erasure coding where data is split and distributed across nodes with redundant information embedded in the encoding. Rather than copying entire data sets, the system creates coded versions where any k segments can reconstruct the original, providing redundancy with minimal additional storage requirements compared to full replication.
3Reliability
If erasure coding is implemented across multiple storage nodes, then data reliability is improved, but computational complexity and processing time worsen
Solution Approach 1:
The patent performs erasure coding operations in advance during data write operations, encoding data into multiple segments and distributing them across storage nodes before any potential failure occurs. This preliminary encoding action ensures that data is immediately protected against failures without requiring real-time computational overhead during read or failure scenarios. The computational work is shifted to the write path, optimizing read performance.
4Adaptability or versatility
If distributed storage architecture is used, then storage scalability is improved, but system complexity and management difficulty worsen
Solution Approach 1:
The patent creates a universal storage pool that can accommodate multiple types of storage nodes and data types through a unified interface and management layer. The distributed storage system provides multi-functional capabilities including block storage, file storage, and object storage through the same underlying infrastructure, reducing management complexity despite architectural diversity. The system abstracts node-specific details behind a common interface, enabling scalable addition of heterogeneous storage resources.
Data Source
AI summary
Sizing a virtual storage system, including: determining a change to one or more resource demands; determining, based on the change to the one or more resource demands, one or more modifications to one or more virtual components included as part of a virtual storage system architecture of a virtual storage system within a cloud computing environment; and initiating, responsive to the change to the one or more resource demands, the one or more modifications to the one or more virtual components included as part of the virtual storage system architecture of the virtual storage system, including replacing one or more of the virtual components with a higher performance virtual component.