Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Master of Science

Program

Computer Science

Supervisor

Sedig, Kamran

2nd Supervisor

Haque, Anwar

Abstract

With the increasing size and complexity of next-generation communication networks, it is critical to utilize interactive information visualization to support the network monitoring, planning, and management. Effectively visualizing large-scale networks has been considered difficult with traditional methods because of the high link density and complicated node relationship. Given the limited screen space, it is essential to explore how to present ultra-large-scale network data efficiently that can assist Internet Service Provider’s (ISP) network planning and management activities. This work proposes a design of the real-time interactive visualization system that combines the idea of progressive disclosure and multiple panels to elegantly visualize the large-scale network and avoid the information-overloaded problem. The system also visualizes the configuration of the network elements and provides the network performance information, including the port-level Quality of Service (QoS) metrics. Additionally, the system enables navigation through the port-level connection and provides different modes for multiple purposes.

Summary for Lay Audience

Conducting network planning and management and monitoring network behaviors have been considered more and more important with the growth in network complexity and volume. The efficient network and service management with the large-scale communication network has been recognized as an essential part of the business goals. The network operators and planners are required to measure and analyze the large-scale network. With the increasing size and complexity of networks, more researchers consider it vital to utilize information interactive visualization tools to encourage data exploration and real-time analysis, and visualization can support timely response and decision making.

Effectively visualizing a moderately large-scale network has been considered difficult with traditional methods because the density of links is high and the relationship between nodes is complicated. Given the limited screen space, it is essential to explore how to manage large quantities of network data efficiently and also contain the necessary network configuration and performance details. Providing performance metrics can benefit the network operators in understanding and monitoring the behavior of the network, and Quality of Service (QoS) management is one of the essential tasks in the network service management. This work proposes a design of the real-time interactive visualization system that combines the idea of progressive disclosure and multiple panels to elegantly visualize the large-scale network according to its hierarchy and avoid the information-overloaded problem. The system also visualizes the configuration and the inside structure of the basic network elements. In addition, the visualization system provides the network performance information including the port-level QoS metrics information, and the system also enables navigation through the port level connection. The above are expected to benefit the network planners/network managers in conducting their network health analysis in a more effective way.

Share

COinS