Distributed Loop Detection Using Node Identifier Propagation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing loop detection methods in distributed systems, such as distributed database systems, rely excessively on a single node for maintaining and transferring global dependency paths, leading to performance deterioration and inability to handle multi-out-degree scenarios, and are prone to failure if the node is killed.
Innovation Solution
A loop detection method where each node maintains a candidate state and public/private identifiers, updating these based on adjacent nodes, allowing for efficient loop detection without relying on a single node, and enabling multi-out-degree scenarios by transferring only valid identifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single node maintains and transfers the global dependency path, then loop detection can be performed, but the node performance deteriorates and the system becomes vulnerable to node failure
Solution Approach 1:
The patent divides the global dependency path maintenance task into segments distributed across multiple nodes. Each node maintains its own local dependency path information rather than relying on a single node to maintain the entire global path. This segmentation distributes the computational burden and eliminates the single point of failure, resolving the contradiction between reliability and node performance.
Solution Approach 2:
The patent introduces a distributed identifier transfer mechanism where nodes exchange dependency path information through standardized messages rather than through a central coordinator. This intermediary message-passing system enables loop detection while distributing the workload, preventing any single node from becoming a performance bottleneck or failure point.
2Adaptability or versatility
If a single node maintains the global dependency path, then loop detection is possible, but the system cannot handle multi-out-degree scenarios
Solution Approach 1:
The patent implements a universal loop detection mechanism that handles both single-out-degree and multi-out-degree scenarios through the same distributed identifier transfer protocol. Each node can independently process dependency paths regardless of how many downstream nodes it has, making the system adaptable to various topological complexities without requiring different management approaches.
Solution Approach 2:
The patent enables dynamic adaptation to changing dependency relationships by allowing nodes to continuously exchange and update their local dependency information. When a node's out-degree changes or new dependency relationships form, the distributed mechanism automatically adjusts through message passing, maintaining correctness without increasing structural complexity.
3Measurement precision
If nodes transfer large amounts of dependency path information, then complete loop detection is achieved, but transaction execution performance deteriorates
Solution Approach 1:
The patent extracts only the essential identifier information needed for loop detection from the complete dependency path data. Instead of transferring entire dependency path structures between nodes, the system exchanges compact identifiers that suffice for detecting cycles. This extraction maintains loop detection accuracy while dramatically reducing the amount of data transferred during transaction execution.
Solution Approach 2:
The patent implements partial information transfer where nodes exchange only the portion of dependency information necessary for loop detection rather than complete path details. This partial action approach provides sufficient precision for the loop detection function while minimizing the performance overhead associated with data transfer and processing.
Data Source
AI summary
This application provides a method including updating a candidate state of the node based on a candidate state of an adjacent downstream node; updating a public identifier of the target node to a public identifier of the adjacent downstream node in response to that the candidate state of the adjacent downstream node is the first candidate state and the public identifier of the target node is less than the public identifier of the adjacent downstream node; and in response to that the candidate state of the adjacent downstream node is the first candidate state and a private identifier of the target node is equal to the public identifier of the adjacent downstream node, determining that a loop exists in a dependency path corresponding to the target node.