Hybrid Server Overload Control via CPU and Buffer Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current server overload control methods rely on either computer processing unit utilization or server buffer occupancy as a measure, failing to effectively manage system overload during periods of high demand, such as media-stimulated events or equipment failures.
Innovation Solution
A method and system that utilize both computer processing unit utilization and server buffer occupancy as measures to determine optimal buffer size and service request limits, dynamically adjusting to maximize processor throughput and maintain system integrity during overload periods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If only computer processing unit utilization is used as a measure of system overload, then the overload control is simple to implement, but the system cannot effectively manage buffer overflow during high demand periods
Solution Approach 1:
The patent combines CPU utilization measurement with buffer occupancy measurement into a unified overload control mechanism. The controller monitors both metrics simultaneously and uses their combination to make admission control decisions, thereby resolving the contradiction between implementation simplicity and effective overload management.
Solution Approach 2:
The overload control system is designed to perform multiple functions: monitoring CPU utilization, monitoring buffer occupancy, calculating optimal buffer size, and making admission control decisions. This multi-functional approach enables the system to effectively manage various aspects of server overload using a single integrated mechanism.
2Reliability
If only server buffer occupancy is used as a measure of system overload, then buffer overflow can be detected, but the system cannot optimize processor throughput during varying utilization conditions
Solution Approach 1:
The system implements feedback by continuously monitoring both buffer occupancy and CPU utilization, then using this information to dynamically calculate optimal buffer size and make admission control decisions. This closed-loop feedback mechanism enables the system to optimize processor throughput while preventing buffer overflow across varying utilization conditions.
Solution Approach 2:
The patent introduces dynamic adjustment of buffer size based on real-time CPU utilization levels. The optimal buffer size is recalculated whenever utilization conditions change, allowing the system to adapt to varying workloads and maximize processor throughput while maintaining reliable buffer overflow protection.
3Device complexity
If static optimal buffer size is used, then the buffer configuration is simple, but the system cannot adapt to changing processing demands and utilization variations
Solution Approach 1:
The patent transforms the static buffer size configuration into a dynamic system that automatically adjusts buffer size based on real-time CPU utilization measurements. The controller continuously monitors utilization and recalculates optimal buffer size, enabling the system to adapt to changing processing demands without manual intervention while maintaining relatively simple operational complexity.
4Quantity of substance
If buffer size is increased to handle high demand, then more requests can be buffered, but processor throughput decreases due to increased waiting time
Solution Approach 1:
The patent dynamically changes the buffer size parameter based on CPU utilization levels. During low utilization periods, larger buffer sizes are permitted to absorb demand surges. During high utilization periods, buffer size is reduced to minimize waiting time and maintain processor throughput. This parameter adjustment resolves the contradiction between buffer capacity and processing efficiency.
Data Source
AI summary
Described are methods and apparatuses, including computer program products, for controlling server resources. An occupancy value of a buffer and a first utilization value of a first processor are measured. The buffer is configured to store one or more requests for service from at least a first client. The first processor is configured to receive and process the one or more requests from the buffer. A buffer size value is determined based on a processor throughput calculation, and a first service request limit for the first client is determined based on the occupancy value, the first utilization value, and the buffer size value.