Distributed Redundant Storage System for Parallel Data Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage systems face inefficiencies in data recovery and durability due to sequential reading from individual storage devices, leading to prolonged repair times and uneven load distribution across storage devices.
Innovation Solution
The system implements a distributed redundant storage approach by selecting storage devices based on pairing values to distribute objects and clusters across multiple devices, allowing for parallel reads and balancing data distribution, thereby reducing repair time and improving throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data objects are stored on individual storage devices with sequential reading, then data recovery can be performed, but repair time is prolonged and system productivity is reduced
Solution Approach 1:
The patent segments data objects into multiple parts and distributes them across multiple storage devices. When a failure occurs, multiple storage devices can be accessed in parallel to recover the segmented data, eliminating the sequential reading bottleneck and reducing repair time while maintaining data recovery capability.
Solution Approach 2:
The patent combines multiple storage devices into a distributed storage system where data objects are replicated across devices. This merging enables parallel read operations during data recovery, improving productivity by reducing repair time while preserving reliability through redundant storage paths.
2Quantity of substance
If data is concentrated on fewer storage devices, then storage capacity is utilized, but load distribution becomes uneven and system performance deteriorates
Solution Approach 1:
The patent implements local quality by distributing data objects and their replicas across multiple storage devices based on local capacity and performance characteristics. This ensures even load distribution while maintaining adequate storage capacity utilization, as each device handles a balanced portion of the total data load.
Solution Approach 2:
The patent introduces a new dimension of distribution by spreading data across multiple storage devices rather than concentrating it. This multi-dimensional distribution approach balances the load across the storage system, improving productivity through better resource utilization while maintaining adequate storage capacity.
3Device complexity
If sequential reading is used from individual storage devices, then system complexity is reduced, but repair time increases and durability is compromised
Solution Approach 1:
The patent segments data recovery operations into parallel reads from multiple storage devices. This segmentation transforms the recovery process from a complex sequential operation to a simpler parallel operation, reducing the perceived complexity while dramatically cutting repair time and improving durability.
Solution Approach 2:
The patent performs preliminary distribution of data objects and their replicas across multiple storage devices during normal operation. This preliminary action ensures that when a failure occurs, the system is already positioned to perform parallel reads, reducing repair time without adding operational complexity during the recovery phase.
4Reliability
If data objects are replicated across multiple storage devices, then durability is improved, but system complexity increases
Solution Approach 1:
The patent implements a universal storage distribution mechanism that automatically manages data replication across multiple devices. This multi-functional system handles both data storage and load balancing through a single distributed architecture, improving durability while avoiding the need for separate complex management systems.
Solution Approach 2:
The distributed storage system performs self-service by automatically distributing data objects and their replicas across storage devices based on available capacity. This self-organizing behavior improves durability through automatic replication while minimizing the complexity of manual management, as the system adapts to changing conditions without external intervention.
Data Source
AI summary
A device implementing a system for object placement includes at least one processor configured to identify an object stored on two of a plurality of storage devices, and to determine a pairing value for each pair of the plurality of storage devices, the pairing value corresponding to a number of stored objects common to the pair. The at least one processor is further configured to select, based on the determined pairing values, two storage devices having a lowest pairing value relative to the determined pairing values, and to move the object to each of the two selected storage devices.