Distributed Storage Node Identification via Geographic Hashing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed storage systems face performance degradation due to increased search paths and latency as more nodes participate, as conventional decentralized solutions fail to account for geographic location correlations, leading to inefficient data sharing and searching across geographically distant nodes.
Innovation Solution
A method that determines node identifiers in a distributed storage system by combining geographic location information and hash values, allowing for logical location determination and efficient data processing requests within regions, thereby optimizing data sharing and searching by reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If more nodes participate in the distributed storage system, then the system capacity and availability are improved, but the search path length and latency increase
Solution Approach 1:
The patent segments the distributed storage system into multiple regions based on geographic location. Each region is further divided into zones, and nodes are organized in a hierarchical structure (region-zone-node) that enables localized data storage and retrieval. This segmentation allows the system to maintain high availability across the entire distributed network while reducing search latency by confining data access to nearby regional zones rather than traversing the entire global node list.
Solution Approach 2:
The patent introduces geographic location information as an intermediary factor in node identification and data routing. By incorporating geographic coordinates into the node identification process and using this information to determine data storage locations, the system creates an intermediary layer that mediates between the global distributed storage requirements and local geographic proximity, enabling efficient data retrieval without requiring traversal of all nodes.
2Device complexity
If conventional decentralized storage mechanisms are used, then system simplicity is maintained, but data sharing efficiency across geographically distant nodes deteriorates
Solution Approach 1:
The patent creates a universal node identification mechanism that simultaneously handles multiple functions: it identifies nodes within the distributed storage system, determines their geographic locations, assigns them to specific regions and zones, and guides data routing decisions. This multi-functional approach maintains architectural simplicity by using a single unified identification scheme rather than requiring separate mechanisms for each function, while significantly improving data sharing efficiency through geographic-aware routing.
3Adaptability or versatility
If nodes are distributed globally without geographic consideration, then system scalability is improved, but search path efficiency deteriorates
Solution Approach 1:
The patent applies local quality by creating geographic zones and regions that are optimized for local data access. Nodes within the same geographic zone are prioritized for data storage and retrieval operations, creating localized optimization while maintaining global scalability. This allows the system to scale globally across different geographic regions while ensuring that data search and access operations proceed efficiently within each local region, avoiding unnecessary traversal of distant nodes.
Data Source
AI summary
Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for data storage. The method includes determining, based on attribute information of each of a set of nodes in a distributed storage system, a hash value of each of the set of nodes; determining a node identifier of each of the set of nodes based on geographic location information of each of the set of nodes and the hash value, with the node identifier being used to indicate a logical location of a node in the distributed storage system; and determining, in response to a processing request from a first node in the set of nodes for first data, a target node in the set of nodes based on the node identifier of each of the set of nodes. The target node is used to process the processing request for the first data.