Adaptive Load Admission Control for Application Server Heap Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory overload protection schemes for application servers, such as those in J2EE environments, struggle to effectively manage memory utilization due to variable workload characteristics, leading to potential crashes and performance degradation, especially in real-time applications like VoIP, as they rely on fixed parameters and fail to adapt to changing conditions.
Innovation Solution
A method and apparatus that detect actual heap utilization and load admission rate, estimate load characteristics variation parameters, and use a feedback controller with adaptive control gain to dynamically regulate the load admission rate, maintaining target heap utilization through a detection module, parameter estimation module, and feedback controller.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fixed parameter memory overload protection schemes are used, then configuration is simple, but the protection cannot adapt to changing workload characteristics and fails to effectively prevent memory overload
Solution Approach 1:
The patent implements dynamic control by continuously monitoring actual heap utilization and load admission rate, then adjusting the load admission rate dynamically based on real-time system state. The controller adapts to changing workload characteristics (call arrival rate, call hold time, memory per call) by modifying control parameters on-the-fly, transforming the static fixed-parameter approach into a dynamic adaptive system that responds to actual system conditions.
Solution Approach 2:
The patent employs feedback control mechanisms where the controller continuously monitors actual heap utilization and compares it against target utilization levels. Based on this feedback, the controller adjusts the load admission rate to maintain optimal memory utilization. The feedback loop incorporates monitoring of workload characteristics and uses this information to adaptively tune control parameters, ensuring the system responds appropriately to changing conditions while preventing memory overload.
2Measurement precision
If fixed control gain feedback controller is used, then implementation is straightforward, but the controller cannot track changes in application characteristics well
Solution Approach 1:
The patent implements adaptive control gain adjustment by continuously monitoring workload characteristics (call arrival rate, call hold time, memory consumption per call) and modifying the control gain parameter accordingly. When workload characteristics change significantly, the controller adjusts the gain to maintain optimal response. This parameter adaptation enables the controller to track application characteristics precisely while managing complexity through structured adaptation rules based on monitored system state.
3Reliability
If explicit configuration of protection parameters is required, then parameter setting is based on experience, but optimal values are hard to determine due to varying application characteristics
Solution Approach 1:
The patent implements self-tuning capability where the controller automatically monitors workload characteristics and adjusts control parameters without requiring manual configuration. The system performs self-diagnosis of memory utilization patterns and self-adjusts the load admission rate and control gain based on observed workload characteristics. This self-service approach eliminates the need for experienced administrators to manually tune parameters, while maintaining reliable memory overload protection through automated adaptation to changing application behavior.
Data Source
AI summary
A method and system for controlling load admission rate of an application server is provided. In the method, actual heap utilization and load admission rate of the application server in current control cycle are detected; based on the detected actual heap utilization and load admission rate, load characteristics variation parameters of the application server are estimated; and control gain of a feedback controller is calculated based on the load characteristics variation parameters; the feedback controller calculates desired load admission rate of the application server in next control cycle by using the calculated control gain. The load admission rate of the application server can be adaptively controlled by tracking changes in workload characteristics of the application server, such that the actual heap utilization of the application server is maintained at or close to target heap utilization.