Weighted Round-Robin Load Balancing Using Performance Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data processing networks, system overload conditions are difficult to manage without adding additional processing units, and existing methods lack efficient load balancing strategies to distribute work requests effectively across distributed processing units based on performance metrics.
Innovation Solution
A method involving obtaining utilization values for each distributed processing unit, applying a mapping function to derive weights, and using these weights for weighted round-robin load balancing, which includes randomizing the distribution list to optimize the distribution of work requests and reduce overload without requiring additional units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If additional data processing units are added to accommodate future demand and reduce system overload, then system capacity and reliability are improved, but system cost and complexity increase
Solution Approach 1:
The load balancing system dynamically adjusts the distribution of work requests to processing units based on real-time performance metrics. The mapping function continuously transforms utilization values into weights, creating a dynamic load distribution mechanism that adapts to changing system conditions without requiring physical system changes
Solution Approach 2:
The system changes the parameter of work request distribution by using a mapping function that transforms utilization values into weights. This parameter transformation enables flexible load balancing where the distribution characteristics can be adjusted through software-based weight adjustments rather than hardware modifications
2Ease of operation
If work requests are distributed evenly across all processing units, then simplicity of load balancing is maintained, but system overload conditions cannot be effectively reduced
Solution Approach 1:
The load balancing system implements feedback by monitoring performance metrics of processing units and using this information to adjust work request distribution. The mapping function takes utilization values as input and produces weights that reflect current system state, creating a closed-loop control mechanism that responds to actual processing unit conditions
Solution Approach 2:
The system performs preliminary action by calculating weights based on performance metrics before distributing work requests. The mapping function pre-processes utilization values into weights, allowing the load balancer to make informed distribution decisions that proactively prevent overload conditions rather than reacting to them
3Reliability
If performance-based weight adjustment is implemented for load balancing, then system overload is reduced, but measurement and calculation complexity increases
Solution Approach 1:
The mapping function serves as an intermediary between performance metrics and load distribution decisions. It transforms raw utilization values into meaningful weights that can be directly used for load balancing, simplifying the overall system architecture by providing a clear translation layer between measurement and action
Data Source
AI summary
Performance parameters are obtained for distributed processing units. The performance parameters include a utilization value, of each distributed processing unit. Respective weights are obtained for the distributed processing units by applying a mapping function to the utilization values. The respective weights are used for weighted round-robin load balancing of work requests upon the distributed processing units. In one implementation, a set of utilization values of the processing units are collected in response to a periodic heartbeat signal, a set of weights are produced from the set of utilization values, a distribution list is produced from the set of weights, and the distribution list is randomized repetitively for re-use during the heartbeat interval.