Load Balancing Feedback Controller with Server State Awareness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current load balancers detect server state changes only through measurable performance changes, leading to slow detection and reaction, resulting in degraded application server performance due to unmanaged state changes.
Innovation Solution
Implementing a load balancing feedback controller enhanced with feed-forward control that adjusts routing weights in response to notifications of impending server state changes, allowing for proactive management of server loads based on current and anticipated states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If load balancers detect server state changes only through measurable performance changes, then the detection mechanism is simple, but the detection speed is slow and performance degradation occurs
Solution Approach 1:
The patent applies preliminary action by having the load balancer receive notification of impending server state changes before the changes actually occur. This allows the load balancer to proactively adjust routing weights and redistribute traffic away from servers experiencing state changes, preventing performance degradation rather than reacting to it after detection. The notification mechanism enables advance awareness of events like application starts, stops, or deployments before they impact server performance metrics.
2Reliability
If the load balancer waits for performance changes to become measurable, then the detection method is passive and simple, but the reaction time is delayed resulting in degraded application server performance
Solution Approach 1:
The patent implements feedback by establishing a notification channel from the server or monitoring system to the load balancer that provides real-time information about impending state changes. This active feedback mechanism allows the load balancer to immediately respond to state changes as they occur, adjusting routing weights dynamically to maintain application server performance. The feedback loop ensures the load balancer has current information about server conditions without waiting for performance degradation to become measurable.
Data Source
AI summary
A computer implemented method, apparatus, and computer usable program code for load balancing in a cluster. In response to receiving a notification of an impending state change event in the server, a weight adjustment value for a server is identified to form an identified weight adjustment. An adjusted routing weight for the server is generated using the identified weight adjustment to form a current routing weight for the server. An incoming service request is routed to a selected server in the cluster using current routing weight associated with servers in the cluster.