Streamed Application Instance Scaling for User-Absence Server Cost

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Cloud gaming systems incur unnecessary server costs when users are away from their seats, particularly due to high-performance server instances that continue to run even when not in use, leading to increased expenses.

Innovation Solution

An electronic device with processors configured to manage multiple performance levels for application execution and streaming, dynamically adjusting instances based on user presence or absence, using subprocessors to reduce performance and costs when the user is away.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a high-performance server instance is maintained to ensure game execution quality, then the application performance is improved, but the server cost increases even when the user is away from their seat

Engineering Contradiction:
Improveapplication performanceVSAvoidserver cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The server instance dynamically adjusts its performance level based on user presence status. When the user is present, the system uses a first instance with first performance level to ensure high-quality game execution. When the user is away, the system switches to a second instance with second performance level (lower than first) to reduce server costs, while maintaining the capability to quickly restore high performance when needed.

Inventive Principle:
Principle #15Dynamics

2Loss of energy

If the server instance is terminated when the user is away from their seat, then the server cost is reduced, but the game playing is terminated regardless of the user's intent

Engineering Contradiction:
Improveserver costVSAvoidgame continuity
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

Instead of completely terminating the server instance when the user is away, the system changes the performance parameter by switching to a second instance with lower performance level. This allows the instance to remain active and maintain game state, ensuring game continuity while reducing server costs. The performance level can be adjusted based on user presence without fully shutting down the service.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If multiple server instances are maintained for multiple users, then the service quality for each user is ensured, but the total server cost increases significantly

Engineering Contradiction:
Improveservice qualityVSAvoidtotal server cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The server instance is designed to serve multiple purposes and adapt to different user presence scenarios. The same instance can operate at different performance levels (first performance level when user is present, second performance level when user is away), making it a universal solution that maintains service quality while adjusting resource consumption based on actual needs.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4697674A1Electronic device for reducing server cost, and control method thereof
Publication Date: 2026.02.18 SAMSUNG ELECTRONICS CO LTD
  • EP4697674A1 patent drawingFigure 1A
  • EP4697674A1 patent drawingFigure 1B
  • EP4697674A1 patent drawingFigure 2

AI summary

An electronic device is disclosed. This electronic device may include a communication interface and at least one processor connected to the communication interface to control the electronic device, wherein the processor executes an application with a first performance by driving a first instance, controls the communication interface to stream a screen corresponding to the execution of the application to an external device, and when information corresponding to user absence is received from the external device through the communication interface, executes the application with a second performance lower than the first performance by driving a second instance.