Proxy Server Resource Scheduling via Request Queue Blocking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional flexible scheduling mechanisms for server resources fail to accurately reflect the true utilization status, leading to inefficiencies and potential system crashes during peak or low demand periods, as they primarily rely on system resource utilization rather than the actual blocking status of application requests.
Innovation Solution
A method and system for resource scheduling that monitors the blocking status of application requests by collecting data from a proxy server, analyzing the number of requests in the queue, and adjusting computing resources based on predetermined scheduling rules to ensure accurate allocation and utilization, including expansion or contraction of resources according to specific conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computing resources are expanded based on system resource utilization (CPU, memory, network flow, magnetic disk IO), then the system can handle peak demand, but resources are wasted under normal conditions and the true utilization status is not reflected
Solution Approach 1:
The patent implements a feedback mechanism by monitoring the blocking status of application requests and using this information to dynamically adjust computing resources. The scheduling mechanism continuously collects blocking status data, analyzes it, and adjusts resource allocation accordingly, creating a closed-loop control system that responds to actual system needs rather than relying on static resource utilization thresholds
Solution Approach 2:
The patent changes the monitoring parameter from system resource utilization (CPU, memory, network flow, magnetic disk IO) to the blocking status of application requests. This parameter change enables more accurate detection of true system utilization needs, as blocking status directly reflects whether applications are actually waiting for resources rather than merely showing resource availability metrics
2Loss of energy
If computing resources are reduced to save costs during low demand, then idle resources are minimized, but the system becomes slow or crashes when peak demand occurs
Solution Approach 1:
The patent implements dynamic resource allocation by continuously monitoring blocking status and adjusting computing resources in real-time based on actual demand. Rather than using static resource allocation or simple threshold-based scaling, the system dynamically adapts resource levels to match the true utilization needs reflected by application request blocking patterns
Solution Approach 2:
The system uses feedback from blocking status monitoring to guide resource allocation decisions, ensuring resources are expanded before blocking becomes severe and contracted only when blocking indicates excess capacity, thereby optimizing both cost efficiency and system performance
3Device complexity
If the scheduling mechanism relies on system resource utilization metrics, then resource allocation is simplified, but it fails to reflect the true status of application processing needs
Solution Approach 1:
The patent changes the monitoring parameter from system resource utilization metrics (CPU, memory, network flow, magnetic disk IO) to the blocking status of application requests. This parameter change provides more accurate measurement of true utilization needs, as blocking status directly indicates whether applications are actually waiting for processing rather than merely showing resource availability
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
An apparatus and method for resource scheduling on an application system utilizing a proxy server, the application system including one or more hosts and running one or more applications, each of the one or more applications having one or more instances. The method includes monitoring a request queue of application requests, at the proxy server, to be processed by the application system for the one or more applications, and scheduling computing resources of the application system for an application of the one or more applications according to a predetermined scheduling rule and a status of the request queue.