Multi-Volume Bit-Stream Storage With Centralized Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current large-scale data storage systems face challenges in achieving desirable error rates and low monetary costs per gigabyte while storing data across multiple storage volumes, as the capacity of individual volumes increases, making it difficult to prevent unrecoverable errors.
Innovation Solution
The method involves segmenting data into multiple segments and transferring them as bit streams across multiple physical storage volumes, where each volume writes the data in its received condition without additional processing, using error-correcting codes like Reed-Solomon and cyclic redundancy checks, and simplifying the storage volumes to only write control and timing bits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is stored across multiple storage volumes using traditional RAID methods, then error protection and access rates are improved, but device complexity and monetary costs increase
Solution Approach 1:
The patent extracts the error correction and data processing functions from individual storage volumes and consolidates them into a centralized storage controller. The storage controller performs all ECC encoding, interleaving, and data segmentation, while storage volumes are simplified to only perform basic write operations, thereby reducing device complexity while maintaining reliability
Solution Approach 2:
The patent merges multiple error correction codes (C1 and C2 Reed-Solomon codes) and data processing operations into a unified error correction system managed by the storage controller. This consolidation reduces the complexity that would arise from implementing multiple separate error correction systems across distributed volumes
2Reliability
If data is stored across multiple storage volumes using traditional RAID methods, then error protection and access rates are improved, but monetary costs per gigabyte increase
Solution Approach 1:
By extracting complex error correction functionality from expensive high-performance storage volumes and placing it in a centralized controller, the system can use cheaper storage volumes that only need basic write capabilities, thereby reducing monetary costs while maintaining error protection
Solution Approach 2:
The patent enables the use of simpler, less expensive storage volumes by offloading the complex error correction work to the storage controller. The storage volumes become simpler components that can be replaced more easily and at lower cost, while the expensive error correction intelligence is centralized
3Reliability
If storage volumes perform full error correction processing, then data protection is improved, but processing speed and storage efficiency decrease
Solution Approach 1:
The patent merges all error correction processing operations into the storage controller, allowing parallel processing of multiple data streams simultaneously. This centralized approach improves productivity by eliminating the need for each storage volume to independently perform error correction, while maintaining comprehensive data protection through the unified ECC system
Data Source
AI summary
Embodiments disclosed herein provide systems, methods, and computer readable media for storing data to a plurality of physical storage volumes. In a particular embodiment, a method provides identifying first data for storage on the plurality of physical storage volumes. Each of the plurality of storage volumes corresponds to respective ones of a plurality of data channels. The method further provides segmenting the first data into a plurality of data segments corresponding to respective ones of the plurality of data channels and transferring the plurality of data segments as respective bit streams over the respective ones of the plurality of data channels to the respective ones of the plurality of physical storage volumes. The plurality of storage volumes stores the respective bit streams in the exact condition in which the bit streams are received.