Decision Scheduling for Cloud Computing Tasks Relying on Solving Large Linear Systems of Equations.

Area of Science:

• Computer Science
• Cloud Computing
• Algorithm Analysis

Background:

• Big Data and cloud computing are increasingly integrated into daily life and work.
• Parallel algorithms are crucial for solving large linear equations in various applications.
• Efficient task scheduling is essential for managing dependent parallel tasks within resource constraints.

Purpose of the Study:

• To propose and summarize a cloud computing task scheduling model.
• To address the efficient execution of mutually dependent parallel tasks.
• To satisfy user expectations regarding task completion time, bandwidth, reliability, and cost.

Main Methods:

• Studied technologies for solving large-scale linear equations.
• Proposed the M-QoS-OCCSM (Multi-Quality of Service-Oriented Cloud Computing Task Scheduling Model).
• Utilized large-scale linear equation solving in task scheduling experiments to evaluate algorithms.

Main Results:

• The M-QoS-OCCSM model efficiently executes N mutually dependent parallel tasks.
• MPQGA algorithm demonstrated faster convergence speeds: 32 seconds (task load 10) and 95 seconds (task load 20) faster than BGA.
• The model effectively balances task execution efficiency with user-defined Quality of Service (QoS) parameters.

Conclusions:

• The M-QoS-OCCSM model offers an effective solution for cloud computing task scheduling.
• The MPQGA algorithm shows superior performance in terms of convergence speed for large-scale task scheduling.
• This research contributes to optimizing parallel task execution in cloud environments.