Dynamic Request Configuration Adjustment via Time-Series Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing systems face challenges in dynamically adjusting request configurations for hardware and software components across multiple geographical regions, particularly due to uncertain parameters such as fluctuating user demand and varying exchange rates, which complicates resource allocation and pricing.
Innovation Solution
A computer-implemented method using machine learning to predict parameter values based on historical time-series data, generating multiple configurations for different time intervals with fixed resource types and uncertainty criteria, allowing users to select and automate adjustments based on satisfied criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple configurations are generated for different time intervals to handle uncertain parameters, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent segments the time interval into multiple sub-intervals and generates separate configurations for each segment. This allows the system to handle uncertain parameters by creating targeted configurations for specific time periods rather than managing a single complex configuration for the entire duration, thus improving adaptability while controlling complexity through structured division.
Solution Approach 2:
The system dynamically adjusts configurations based on predicted parameter values for different time intervals. By making configurations time-dependent and adaptable rather than static, the system improves its ability to respond to changing conditions while the automated selection process manages the complexity of having multiple configurations.
2Measurement precision
If machine learning processes are used to predict parameter values, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary predictions of parameter values using machine learning processes before final configuration selection. By predicting parameter values in advance for different time intervals, the system improves measurement precision of uncertain parameters while the automated framework manages the complexity of integrating these prediction processes into the overall configuration system.
3Productivity
If automated actions are initiated based on uncertainty criteria, then productivity is improved, but device complexity increases
Solution Approach 1:
The system implements self-service through automated actions that are automatically initiated when uncertainty criteria are satisfied. The configuration management system monitors parameters and automatically selects and applies appropriate configurations without manual intervention, improving productivity while the automated decision-making framework manages the complexity of the monitoring and selection processes.
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for dynamic adjustment of request configurations are provided herein. An example method includes: determining that a value of a first parameter associated with a request varies over a particular period of time; predicting the value of the first parameter over the period of time based on historical time-series data; generating configurations for the request based on the predicted value of the first parameter, where each configuration corresponds to a different time interval within the period of time and includes: a second parameter, associated with a type of resource, that is fixed over the corresponding time interval and one or more uncertainty criteria corresponding to the second parameter; obtaining a selection of one of the configurations; and initiating one or more automated actions in response to at least one of the uncertainty criteria of the selected configuration being satisfied during the corresponding time interval.