Load Distribution Unit Scheduling Terminal Requests

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for load distribution between servers and terminal devices in communication networks often lead to server and network overload, resulting in system failures due to inefficient resource utilization and lack of effective load balancing.

Innovation Solution

A method where the load distribution unit analyzes resource utilization data to schedule load processes within specific time windows, allowing terminals to repeat requests in later time intervals, thereby preventing overload by optimizing CPU and network resource usage through a computer algorithm and administrator decision-making.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If terminals request load processes continuously without load distribution control, then immediate service response is achieved, but server and network overload occurs leading to system failures

Engineering Contradiction:
Improveservice response speedVSAvoidsystem reliability
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The load distribution unit performs preliminary analysis of resource utilization data before allowing load process requests to proceed. It proactively identifies periods of high server or network utilization and prevents requests from being executed during these critical periods, thereby avoiding overload while still maintaining system reliability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors resource utilization data from the server and network, and the load distribution unit uses this feedback to dynamically control whether terminals can request load processes. When utilization exceeds thresholds, the system feedback signals terminals to delay requests, creating a self-regulating mechanism that prevents overload.

Inventive Principle:
Principle #23Feedback

2Reliability

If load processes are distributed uniformly over time, then resource utilization is optimized and overload prevented, but service response time increases due to request delays

Engineering Contradiction:
Improvesystem stabilityVSAvoidservice response time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The load distribution control is dynamic rather than static. The system adjusts the strictness of load distribution enforcement based on current resource utilization conditions. When resources are available, requests are processed promptly; when utilization is high, requests are delayed. This dynamic approach balances system stability with acceptable response times.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the parameter of request processing timing based on resource utilization conditions. Instead of uniformly delaying all requests, it selectively delays requests only when resource utilization parameters indicate potential overload, allowing rapid processing when resources are abundant and thus minimizing overall response time loss.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If multiple terminals request load processes simultaneously, then high productivity is achieved, but server and network overload occurs

Engineering Contradiction:
Improvedata processing throughputVSAvoidsystem stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system allows partial simultaneous processing of load processes by multiple terminals, but applies selective control based on current resource capacity. Instead of preventing all simultaneous requests or allowing all requests without limit, it permits a controlled number of concurrent processes that match available server and network capacity, thus maintaining high productivity while preventing overload.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3276897B1Load sharing within a communication network
Publication Date: 2021.01.27 DEUTSCHE TELEKOM AG
  • EP3276897B1 patent drawingFigure 1
  • EP3276897B1 patent drawingFigure 2
  • EP3276897B1 patent drawingFigure 3

AI summary

Method and system for load balancing between a server (24) and a terminal device (14a, b, c, d), wherein the server (24) exchanges data with the terminal device (14a, b, c, d) via a communication network (20), in particular the Internet, wherein a load balancing unit (26) controls the data traffic between the server (20) and the terminal device (14a, b, c, d), wherein the terminal device (14a, b, c, d) requests the download of data from the server (24) at a specific time, wherein the load balancing unit (26) sends the terminal device (14a, b, c, d) information requesting the terminal device to repeat the request at a later time window, and wherein the terminal device (14a, b, c, d) resends his request in the later time window.