On delay adjustment for dynamic load balancing in distributed virtual environments.

Area of Science:

• Computer Science
• Network Engineering

Background:

• Distributed virtual environments (DVEs) are increasingly popular, particularly for massive multiplayer online games (MMOGs).
• Scalability is a critical challenge for interactive DVEs as user numbers grow, often addressed by multi-server architectures.
• Existing load balancing methods for DVEs often overlook the impact of network delay between servers on solution accuracy.

Purpose of the Study:

• To formally analyze the effect of network delay on load balancing accuracy in DVEs.
• To propose and evaluate efficient delay adjustment schemes for improved DVE performance.

Main Methods:

• Formal analysis of network delay's impact on server load distribution.
• Development of two novel delay adjustment schemes.
• Experimental evaluation of the proposed schemes' effectiveness.

Main Results:

• Network delay significantly affects the performance of load balancing algorithms in DVEs.
• The proposed delay adjustment schemes demonstrably improve load balancing performance.
• The schemes achieve significant performance gains with negligible computational overhead.

Conclusions:

• Addressing network delay is crucial for accurate and effective load balancing in scalable DVEs.
• The developed delay adjustment schemes offer a practical solution to enhance DVE performance.
• This research contributes to the design of more robust and scalable interactive DVE systems.