Distributed Storage Node Selection via Multicast Query
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed data storage systems face challenges in achieving simple, robust, and reliable data writing across multiple geographically disparate storage nodes, particularly in selecting appropriate nodes based on data properties and node characteristics.
Innovation Solution
A method and device for writing data in a distributed data storage system that involves sending a multicast storage query to multiple nodes, selecting nodes based on their properties and data properties using an algorithm that calculates probability factors for node selection, and sending data to the chosen nodes, ensuring content-sensitive and reliable storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is written to multiple storage nodes in a distributed system, then data reliability and availability are improved, but the complexity of selecting appropriate nodes and managing the writing process increases
Solution Approach 1:
The patent segments the storage system into multiple independent storage nodes, each capable of storing data independently. The writing process is segmented into parallel operations where data is simultaneously sent to multiple selected nodes, improving reliability while distributing the complexity across independent units.
Solution Approach 2:
The patent implements a feedback mechanism where storage nodes respond to write requests with their availability status and performance characteristics. The server uses this feedback information to dynamically select appropriate storage nodes based on real-time system state, simplifying the decision-making process while maintaining high reliability.
2Productivity
If storage nodes are selected based on multiple properties (response time, redundancy, energy consumption), then data storage optimization is improved, but the selection process complexity increases
Solution Approach 1:
The patent changes the selection parameters dynamically based on data characteristics and system state. Instead of using fixed selection criteria, the system adjusts parameters such as response time thresholds, redundancy requirements, and energy consumption weights according to the specific write operation requirements, optimizing storage while managing complexity through adaptive parameter adjustment.
Solution Approach 2:
The patent creates a universal selection mechanism that can handle multiple node properties and data requirements through a single integrated algorithm. The server uses a unified approach to evaluate nodes based on various properties (response time, redundancy, energy consumption) and selects appropriate nodes for different data types and access patterns, simplifying the selection process while achieving comprehensive optimization.
3Measurement precision
If a multicast query is sent to all storage nodes to gather responses, then node selection accuracy is improved, but network traffic and processing overhead increase
Solution Approach 1:
The patent applies partial action by sending multicast queries to all storage nodes but only processing and selecting from a subset of responsive nodes. Not all nodes need to respond fully - only those that are available and meet basic criteria participate in the selection process, reducing unnecessary network traffic and processing overhead while maintaining sufficient selection accuracy.
Data Source
AI summary
There is disclosed a method for writing data in a data storage system comprising a plurality of data storage nodes, the method being employed in a server running an application which accesses data in the data storage system, and comprising: sending a multicast storage query to a plurality of said storage nodes; receiving a plurality of responses from a subset of said storage nodes, said responses including a storage node property; selecting at least two storage nodes in the subset for storing said data, based on said responses, wherein the selecting is based on a data property of the data to be stored and a storage node property.