Multi-Dimensional Disk Arrays for Video Surveillance Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional redundant data storage systems, such as distributed RAID, face performance issues due to network traffic and are not cost-effective for applications like video surveillance, especially when handling disk failures, which require extensive data movement and rebuild processes.
Innovation Solution
The implementation of multi-dimensional disk arrays with localized fault tolerance, using single-port disk drives and cross-node managers to create redundant storage schemes at multiple levels, allowing for efficient data recovery without relying heavily on network resources, and employing online data migration and table-based storage object displays to manage and rebuild data efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If distributed RAID is used to distribute parity information across multiple nodes, then fault tolerance against disk failures is improved, but system performance deteriorates due to network traffic overhead
Solution Approach 1:
The patent segments the storage system into multiple independent nodes, each maintaining its own local RAID arrays. Parity information is distributed across these segmented nodes, allowing fault tolerance to be achieved without requiring extensive network communication for routine operations. Each node can independently manage its own disk failures using local resources.
Solution Approach 2:
The patent implements local quality by enabling each node to maintain redundant storage capacity locally rather than relying on remote nodes. When a disk fails, the parity information can be reconstructed from local disks within the same node, eliminating the need to fetch data over the network and thus preserving system performance while maintaining fault tolerance.
2Reliability
If disk failure rebuild is performed over the network in distributed RAID, then data recovery is achieved, but data movement time increases significantly
Solution Approach 1:
The patent applies preliminary action by pre-distributing parity information across multiple local disks within each node before any failure occurs. This allows the system to immediately reconstruct lost data from locally available parity disks without needing to retrieve data over the network, dramatically reducing rebuild time while ensuring data recovery capability.
3Reliability
If redundant data-paths with dual-port disk drives and dual controllers are implemented, then protection against component failures is improved, but system cost increases
Solution Approach 1:
The patent merges the control functions into a single controller per node while using software-based RAID management. This consolidation eliminates the need for dual redundant controllers and dual-port disk drives, reducing hardware costs while maintaining fault tolerance through intelligent software management that can handle component failures and initiate rebuilds using locally distributed parity information.
Data Source
AI summary
Embodiments of the present invention generally provide for multi-dimensional disk arrays and methods for managing same and can be used in video surveillance systems for the management of real-time video data, image data, or combinations thereof.