Service Instance Control via User Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In scale-out operations based on service load, increases in service load are addressed after the fact, leading to temporary service delays. The goal is to develop a technique that prevents these delays.

Innovation Solution

A control apparatus and method that predict the number of users of a service before it starts and adjust the instances of the application or service chaining accordingly, ensuring optimal resource allocation and service performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If scale-out is performed according to service load after the load increase occurs, then the service can handle increased user demand, but service delays occur temporarily during the load increase period

Engineering Contradiction:

Improveservice load handling capabilityVSAvoidservice delay time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by predicting the number of users before service provision starts and proactively adjusting the number of application instances in advance. The control apparatus predicts user numbers based on historical data and service information, then scales out instances before the actual load increase occurs, eliminating service delays while maintaining adaptability to demand

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by continuously monitoring service load and user numbers, then using this information to dynamically adjust the number of application instances. The control apparatus receives service information, predicts user demand, and automatically scales resources based on predicted and actual load conditions, creating a closed-loop system that adapts to changing demands

Inventive Principle:

Principle #23Feedback

2Loss of time

If the number of application instances is increased in advance based on predicted user numbers, then service delays are prevented, but resource waste may occur if prediction is inaccurate

Engineering Contradiction:

Improveservice delay preventionVSAvoidresource waste

Core Design Contradiction:

Loss of timeVSLoss of energy

Solution Approach 1:

The patent applies partial action by scaling out application instances partially based on predicted user numbers rather than fully provisioning for maximum potential demand. The control apparatus adjusts instances proportionally to the predicted load increase, ensuring sufficient capacity to prevent delays while avoiding excessive resource allocation that would cause waste

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent uses parameter changes by dynamically adjusting the number of application instances based on changing user number predictions. The control apparatus modifies instance count as a variable parameter in response to predicted demand changes, allowing flexible resource allocation that prevents both service delays and resource waste through adaptive parameter tuning

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250199862A1Control apparatus, control method, and non-transitory computer-readable medium

Publication Date: 2025.06.19 NEC CORP

AI summary

A control apparatus predicts, for a target service to be provided at a service provision location, the number of users of the target service, before the provision starts. The control apparatus performs control regarding an instance of an application that provides the target service, or control of a service chaining among a plurality of the applications, based on the predicted number of users.