Application Load Forecasting via Multi-Server Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current enterprise and network monitoring systems fail to effectively monitor computational resources across multiple servers, making it difficult to identify the primary cause of increased load and forecast future requirements, especially when applications span multiple devices.
Innovation Solution
A method and device that monitor computational load characteristics across multiple processes executed on different computing devices, establish a mathematical relationship between input load and computational load, and forecast future needs using an application poller, aggregator, correlating engine, and enterprise computing engine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional monitoring systems are used to monitor computational resources, then monitoring of single device or similar indicators on multiple devices is achieved, but the ability to track indicia based on application across multiple servers is lost
Solution Approach 1:
The monitoring system segments computational resources by application rather than by device, creating application-specific views of computational indicia across multiple servers. This allows precise tracking of application-based metrics while maintaining manageable system complexity through modular architecture.
Solution Approach 2:
The system introduces an intermediary layer that correlates computational indicia from multiple devices with application identifiers. This intermediary enables application-based tracking without requiring direct complex connections between all monitoring points, thus maintaining system manageability.
2Adaptability or versatility
If monitoring is implemented across multiple servers to track application processes, then application-based tracking capability is improved, but the difficulty of pinpointing the primary cause of load increases
Solution Approach 1:
The system implements feedback mechanisms that provide application-specific computational indicia back to the monitoring interface. This feedback loop enables precise identification of which application is causing increased load, even when processes are distributed across multiple servers, by correlating indicia with application identifiers.
Solution Approach 2:
The system extracts application-specific computational indicia from the aggregate data across multiple servers. By separating and isolating application-related metrics from overall system metrics, the system makes it easier to identify the primary cause of load increases without being overwhelmed by multi-server complexity.
3Productivity
If computational resources are monitored until usage exceeds acceptable levels, then resource utilization is maximized, but the ability to forecast future needs is lost
Solution Approach 1:
The system performs preliminary actions by collecting and analyzing computational indicia data before resource thresholds are exceeded. This preliminary data collection and analysis enables forecasting of future computational needs, allowing proactive resource allocation rather than reactive responses after thresholds are breached.
Solution Approach 2:
The system implements partial monitoring and analysis actions continuously at below-threshold levels. By performing lighter-weight monitoring and analytical operations even when resources are under-utilized, the system accumulates data needed for accurate forecasting without significantly impacting current resource utilization efficiency.
Data Source
AI summary
The invention comprises methods and devices of forecasting future computational needs of an application based on input load, where the application comprises a plurality of processes executed on a plurality of computing devices. The method of the invention proceeds by monitoring at least a computational load characteristic of at least a first process executed on a first computing device and a second process executed on a second computing device. A mathematical relationship between input load and the computational load characteristic is established, and future computational needs are forecasted based on the established mathematical relationship.