Distributed Network Node Load Balancing via Query Frequency Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Decentralized data networks face instability and inefficiency due to 'hot spots' where specific resources are frequently requested, leading to network node overload and increased traffic, with existing solutions like data replication and additional logical networks requiring significant bandwidth and management efforts.
Innovation Solution
The method involves monitoring query frequencies and redistributing table entries along a query path when a threshold is exceeded, using a distributed hash table to assign query values, and implementing reverse propagation and caching to distribute the load and reduce administrative effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is replicated on one or more network nodes to solve the hot spot problem, then the overload of network nodes is reduced, but the network requires further special mechanisms to ensure replicated copies are up-to-date, increasing device complexity
Solution Approach 1:
The patent applies preliminary action by proactively detecting query frequency trends and redistributing table entries before network nodes become overloaded. The central coordinating node monitors query frequencies and performs preventive redistribution when threshold values are approached, avoiding the need for reactive replication mechanisms and their associated complexity.
Solution Approach 2:
The patent introduces a central coordinating node as an intermediary that manages the redistribution of table entries. This intermediary coordinates the load balancing process, allowing network nodes to redirect queries without requiring complex peer-to-peer replication management mechanisms. The intermediary handles the complexity centrally while maintaining simple node behavior.
2Reliability
If a further logical network is provided to organize requests, then the hot spot problem is addressed, but a relatively high bandwidth of transmission capacity is used solely by the further logical network due to keep alive messages
Solution Approach 1:
The patent extracts the coordination function from the data network itself and places it in a separate central coordinating node. This separation allows the data network to focus on data transmission while the coordinating node handles organization and load balancing, eliminating the need for keep-alive messages in the data network and reducing bandwidth consumption.
Solution Approach 2:
The central coordinating node serves as an intermediary that organizes requests and manages load distribution without requiring continuous communication channels. Unlike a persistent logical network that needs keep-alive messages, the intermediary operates on-demand, coordinating redistributions only when necessary based on monitored query frequencies.
3Productivity
If table entries are redistributed along the query path when query frequency exceeds threshold, then network node overload is prevented, but the system requires monitoring and threshold management mechanisms
Solution Approach 1:
The patent implements self-service by enabling network nodes to autonomously monitor their own query frequencies and report to the central coordinating node. Each node independently tracks its load and triggers redistribution requests when thresholds are exceeded, eliminating the need for complex centralized monitoring of individual node states while maintaining simple, uniform node behavior.
Solution Approach 2:
The patent uses parameter changes by monitoring query frequency as a key metric and triggering redistribution when this parameter exceeds a threshold. The system dynamically adjusts the distribution of table entries based on changing query patterns, allowing the network to adapt to varying loads while maintaining simple threshold-based decision logic at each node.
Data Source
AI summary
The invention describes a method for operating a local area data network comprising a plurality of network nodes (Ni), where the network nodes (Ni) can send requests to the data network using request values (Ki) and each of the network nodes (Ni) has an associated explicit identifier from a key space. The inventive method comprises the following steps: a) a plurality of network nodes (Ni) respectively have an associated one or more table entries in a value table distributed over the network nodes (Ni). In this case, each table entry comprises a request value (Ki) comprising data associated with the key space and with the request value (Ki), the network node (Ni) being responsible for a request value (Ki) whose identifier is less than or equal to the request value (Ki); b) for the plurality of network nodes (Ni), the frequency with which requests for the request values associated with a respective network node (Ni) are sent to the respective network node (Ni) is respectively monitored; c) if step b) establishes that the frequency of the requests for one or more request values to a respective network node (Ni) exceeds a predetermined threshold value then at least the table entry or entries for this or these request values are assigned to at least one other network node (Ni) which is situated in a request path (PF) for the request.