Communication Element Scale-Out Using Overlapping Load Predictions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in scaling out elements in a timely and accurate manner due to low-accuracy network load predictions, leading to potential over-scaling or under-scaling issues.
Innovation Solution
A scale-out execution system that predicts network load based on performance index values before a reference time point, determines if the predicted load meets a given condition across multiple overlapping prediction periods, and executes scale-out only when a predetermined number of determinations satisfy the condition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If scale-out is executed early based on low-accuracy prediction to have a safety margin, then the network load increase risk is reduced, but unnecessary scale-out actions increase
Solution Approach 1:
The system performs preliminary scale-out actions based on predicted network load before the actual load increase occurs. By using prediction algorithms to forecast future network conditions, the system proactively scales out elements in advance, ensuring capacity is available when needed while attempting to avoid premature actions through predictive accuracy.
Solution Approach 2:
The system implements a feedback mechanism where actual network load measurements are compared with predicted values. This feedback loop allows the system to learn from prediction accuracy and adjust future scale-out decisions, reducing unnecessary scale-out actions while maintaining reliable network load handling.
2Loss of energy
If scale-out is executed based on high-accuracy prediction, then unnecessary scale-out is reduced, but the response time to actual network load increase may be delayed
Solution Approach 1:
The system dynamically adjusts the prediction horizon and scale-out timing based on current network conditions and prediction confidence levels. When prediction accuracy is high and conditions are stable, it uses longer prediction horizons to optimize resource efficiency. When volatility increases or confidence decreases, it shortens the horizon to improve response time, creating a dynamic balance between efficiency and responsiveness.
3Measurement precision
If multiple overlapping prediction periods are used to determine scale-out, then prediction accuracy is improved, but system complexity increases
Solution Approach 1:
The system segments the prediction process into multiple overlapping time periods, each with its own prediction model and confidence assessment. By dividing the prediction task into manageable segments that overlap in time, the system achieves more accurate forecasting through multiple perspectives while keeping each individual prediction model relatively simple and manageable.
Data Source
AI summary
It is enabled that an element included in a communication system is appropriately scaled out. An AI (70) predicts, based on a performance index value before a given reference time point relating to a communication system, a network load in a prediction period from the given reference time point until a predetermined time after the given reference time point. A policy manager (90) determines, in response to the prediction, whether a magnitude of a predicted value of the network load in the prediction period satisfies a given condition. The policy manager (90), a life cycle manager (94), a container manager (78), and a configuration manager (76) execute scale-out of an element included in the communication system when, in the determinations for two or more prediction periods which have different reference time points but at least partially overlap, a predetermined number or more of the determinations are a determination that the condition is satisfied.