Distributed XOR Parity in Storage Controllers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional RAID systems rely on centralized CPU or GPU for XOR parity calculations, leading to CPU bottlenecks, high costs, and increased power consumption, making them inefficient and complex for data storage systems.
Innovation Solution
A data storage system using distributed XOR calculations across a storage area network with embedded XOR capability in storage controllers, leveraging FPGA technology and PCIe switches for parallel parity generation without the need for a centralized RAID engine or processor, allowing direct routing of read/write operations and efficient scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized CPU or GPU is used for XOR parity calculations, then data protection is provided, but CPU bottlenecks and high costs occur
Solution Approach 1:
The patent divides the centralized XOR calculation function into distributed storage controllers. Each storage controller independently performs XOR operations on its assigned data blocks, eliminating the single-point bottleneck of centralized CPU/GPU processing while maintaining data protection through distributed parity generation.
Solution Approach 2:
The patent extracts the XOR calculation capability from the centralized CPU/GPU and embeds it directly into the storage controllers. This extraction removes the dependency on high-performance centralized processors, reducing costs and eliminating CPU bottlenecks while preserving the data protection function.
2Reliability
If centralized RAID engine is used for XOR operations, then parity data is generated, but high power consumption and cost occur
Solution Approach 1:
Storage controllers perform XOR parity calculations autonomously using their embedded XOR engines, without requiring a centralized RAID engine. This self-service approach eliminates the power consumption and cost overhead of maintaining a dedicated centralized processing unit while ensuring reliable parity data generation.
Solution Approach 2:
The patent removes the centralized RAID engine from the system architecture and distributes its XOR calculation functionality to individual storage controllers. This extraction eliminates the high power consumption associated with centralized processing while maintaining the essential parity generation capability.
3Productivity
If distributed XOR capability is embedded in storage controllers, then scaling is improved, but device complexity increases
Solution Approach 1:
Storage controllers are designed with universal XOR engines that can handle parity calculations for multiple data blocks simultaneously. This multi-functionality allows a single controller to perform distributed XOR operations across different data sets, enabling system scaling without proportionally increasing overall device complexity.
Solution Approach 2:
The patent changes the architectural parameter from centralized to distributed XOR processing. By embedding XOR capability at the controller level rather than requiring a centralized engine, the system achieves linear scalability with storage capacity while the complexity increase per controller remains manageable and standardized.
Data Source
AI summary
In the data storage system the storage area network performs XOR operations on incoming data for parity generation without buffering data through a centralized RAID engine or processor. The hardware for calculating the XOR data is distributed to incrementally calculate data parity in parallel across each data channel and may be implemented as a set of FPGAs with low bandwidths to efficiently scale as the amount of storage memory increases. A host adaptively appoints data storage controllers in the storage area network to perform XOR parity operations on data passing therethrough. The system provides data migration and parity generation in a simple and effective matter and attains a reduction in cost and power consumption.