Software Workload Boundary Forecasting via Resource Tree Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional monitoring tools cannot effectively determine workload boundaries for resource trees spanning across multiple solution layers, nor can they identify specific changes required to accommodate increased throughput without sacrificing performance of other request types.
Innovation Solution
A system and method that forecast and expand software workload boundaries by analyzing resource tree data, sequence information, historical usage, and current throughput values to determine additional potential throughput and necessary adjustments, allowing for increased throughput while maintaining performance across multiple solution layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional monitoring tools are used to detect resource usage thresholds, then resource usage can be monitored, but workload boundaries for multi-layer resource trees cannot be effectively determined
Solution Approach 1:
The patent segments the resource monitoring system into hierarchical layers (resource tree structure) where each layer represents a different level of resource abstraction. The workload boundary determination is performed separately for each layer and request type, allowing precise measurement at multiple granularities while maintaining adaptability across the entire resource hierarchy.
Solution Approach 2:
The patent introduces a new dimension of analysis by considering multiple request types simultaneously and their sequential resource usage patterns. Instead of monitoring resources in isolation, the system analyzes the sequence in which different request types consume resources across layers, enabling effective workload boundary determination for complex multi-layer resource trees.
2Productivity
If resource usage is increased to accommodate higher throughput, then productivity improves, but performance of other request types deteriorates
Solution Approach 1:
The patent performs preliminary analysis of resource sequences and historical usage patterns to forecast workload boundaries before actually increasing throughput. By determining the maximum safe throughput increase in advance (additional potential throughput value), the system can expand capacity while pre-ensuring that performance requirements for other request types are maintained, avoiding the trade-off between productivity and reliability.
3Loss of information
If monitoring tools provide basic threshold alerts, then simple resource usage can be detected, but specific changes required to accommodate increased throughput cannot be identified
Solution Approach 1:
The patent implements a feedback mechanism that analyzes historical usage data and sequence information to provide actionable insights about workload boundaries. The system continuously monitors resource usage patterns, compares them against forecasted boundaries, and provides specific guidance on how much throughput can be safely increased. This feedback loop eliminates the need for complex manual analysis while delivering precise actionable information.
Data Source
AI summary
The present invention is a system and method for forecasting and expanding software workload boundaries. The system includes a computer processor having a non-transitory memory containing program code for receiving a resource tree data set identifying a plurality of resources in a resource tree; receiving a sequence information set indicative of an order of resources for a request type; receiving a historical usage information set indicative of actual resource usage for the request type with respect resources in the resource tree; receiving a current throughput value for the request type; and determining, based at least in part upon the resource data tree set, the sequence information set and the current throughput values, a first additional potential throughput value corresponding to additional throughput with respect to the request type that can be performed by the resource tree in addition to its current throughput.