Distributed Storage Path Cost Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale distributed storage systems face challenges in efficiently allocating storage resources due to increasing redundancy and complexity, requiring dynamic selection of storage paths based on varying costs and constraints, while ensuring data redundancy and reliability.
Innovation Solution
A system that includes a block spreader, storage path estimator, and storage path selector to distribute and retrieve data blocks based on calculated system cost values, using a hierarchical addressing scheme and considering processing, memory, latency, and network parameters, to optimize storage path selection and resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the number of storage elements in a distributed storage system increases to store large amounts of data, then the storage capacity and redundancy improve, but the probability of failure of storage elements and controller components increases
Solution Approach 1:
The patent segments data into multiple data blocks and distributes them across different storage elements. Each data block can be independently stored and retrieved, allowing the system to scale storage capacity while maintaining reliability through distributed redundancy. The segmentation enables parallel access paths, reducing the impact of individual component failures.
Solution Approach 2:
The system dynamically changes parameters such as path cost values, which represent the cost of accessing data through different storage paths. By monitoring system conditions and adjusting path cost values in real-time, the system adapts to changing reliability conditions, routing data access away from paths with higher failure probabilities while maintaining optimal performance.
2Reliability
If redundancy and complexity in distributed storage systems increase to ensure data reliability, then data protection improves, but the challenge of allocating storage resources including processing, memory, and network resources increases
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring path cost values and system conditions. The controller receives feedback about the state of storage paths and adjusts resource allocation decisions based on this feedback. This dynamic feedback loop simplifies resource allocation by providing real-time guidance on optimal paths without requiring complex manual configuration.
Solution Approach 2:
The patent introduces dynamic path cost values that change based on current system conditions such as load, service interruptions, and storage management activities. This dynamic approach allows the system to automatically adapt resource allocation to changing conditions, reducing the complexity of static resource management while maintaining high data protection levels through flexible path selection.
3Productivity
If the system dynamically selects storage paths based on varying costs and constraints to optimize resource allocation, then storage efficiency improves, but the complexity of calculating and comparing system cost values for each storage element increases
Solution Approach 1:
The system performs preliminary calculations of path cost values and stores them for future reference. By pre-calculating and caching cost information, the system reduces the complexity of real-time decision-making while maintaining high storage efficiency. The preliminary action allows the controller to quickly retrieve and compare pre-computed cost values without performing complex calculations during data access operations.
4Measurement precision
If the system uses hierarchical addressing and calculates system cost values based on multiple parameters to select optimal storage paths, then path selection accuracy improves, but the time and computational resources required for path selection increase
Solution Approach 1:
The system applies partial action by considering only the most relevant parameters for path selection in given situations, rather than always evaluating all possible parameters. This selective approach maintains high path selection accuracy when needed while reducing computational overhead and time requirements for routine operations. The system can adjust the level of analysis based on the specific context and requirements.
Data Source
AI summary
Example distributed storage systems, controller nodes, and methods provide distributed and redundant data blocks accessed based on storage path cost values. Storage elements are accessible through hierarchical storage paths traversing multiple system components. Data blocks are distributed among the storage elements. System costs are calculated based on the storage path for reaching each storage element and a storage path is selected based on a comparison of the system costs for each storage element. Data blocks are accessed through the selected storage path.