Distributed Task Scheduling with Dynamic Server Load Thresholds

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing load balancing technologies in cloud platforms fail to meet the diverse needs of tasks due to high dynamics and heterogeneity, resulting in unreasonable task allocation and low scheduling efficiency.

Innovation Solution

A distributed task load balancing method that utilizes performance requirement indicators, historical load change rates, and real-time performance indicators to allocate tasks to server nodes, employing radar charts and deep neural networks for precise node selection and load balancing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If existing load balancing technologies are used, then basic task distribution is achieved, but task allocation becomes unreasonable and scheduling efficiency decreases due to high dynamics and heterogeneity of cloud platform environments

Engineering Contradiction:

Improvescheduling efficiencyVSAvoidadaptability to diverse task needs

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent applies local quality by creating customized matching strategies for different task types. Each task category (CPU-intensive, I/O-intensive, network-intensive, memory-intensive) has its own specific matching algorithm that considers relevant performance indicators, allowing the system to adapt locally to the specific needs of each task rather than using a uniform approach

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent implements dynamics by continuously monitoring real-time performance indicators of server nodes and dynamically adjusting task allocation decisions. The system updates load thresholds based on historical load change rates and recalculates matching degrees as performance conditions change, enabling the load balancing mechanism to adapt to the high dynamics of cloud platform environments

Inventive Principle:

Principle #15Dynamics

2Adaptability or versatility

If traditional load balancing methods are applied, then simple task distribution is achieved, but the system cannot meet diverse resource requirements of different tasks

Engineering Contradiction:

Improveability to meet diverse task requirementsVSAvoidcomplexity of scheduling system

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the complex scheduling problem into distinct task categories (CPU-intensive, I/O-intensive, network-intensive, memory-intensive) and applies specialized matching strategies to each segment. This segmentation allows the system to handle diverse requirements through modular, category-specific approaches rather than attempting a single complex universal solution

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes parameters by selecting different performance indicators and weighting schemes based on task type. For example, CPU-intensive tasks prioritize CPU performance indicators while I/O-intensive tasks prioritize I/O performance indicators. This parameter adaptation enables the system to meet diverse requirements by adjusting which parameters are most important for each task category

Inventive Principle:

Principle #35Parameter changes

3Ease of operation

If static task allocation is used, then system simplicity is maintained, but task allocation becomes unreasonable under dynamic cloud platform conditions

Engineering Contradiction:

Improvesimplicity of allocation mechanismVSAvoidreasonableness of task allocation

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent implements feedback by continuously monitoring real-time performance indicators of server nodes and using this information to make dynamic allocation decisions. The system calculates matching degrees based on current node performance and historical load change rates, then adjusts task allocation accordingly. This feedback loop ensures allocation remains reasonable despite dynamic conditions while maintaining operational simplicity through automated decision-making

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20260086872A1Distributed task load balancing scheduling method and system

Publication Date: 2026.03.26 HUANENG INFORMATION TECH CO LTD

AI summary

Provided are a distributed task load balancing scheduling method and system. The method includes: acquiring performance requirement indicators of tasks to be executed, and allocating the tasks to be executed to server nodes in a server cluster network topology according to the performance requirement indicators; acquiring historical load change rates of the server nodes, and determining a load threshold corresponding to each server node according to the historical load change rates; and acquiring real-time performance indicators of the server nodes, determining load score values according to the real-time performance indicators of the server nodes, screening out overloaded server nodes whose load score values exceed the load thresholds, and performing load balancing scheduling on tasks in the overloaded server nodes. The present disclosure can accurately calculate the optimal allocation node for each task, provide a refined task scheduling strategy for a server cluster network, and effectively improve the load balancing degree.