Back to student theses
Bachelor's Thesis
Infrastructure-Aware Traffic Matrix Generation for Optical Core Networks
Traffic matrices are a fundamental input for the study of optical core networks, since they determine how demands are allocated across the network and strongly influence the outcome of simulation and planning studies. However, many traffic matrices are still built on generic assumptions that are only loosely connected to the actual structure of the network. When information about Central Offices (COs), Data Centers (DCs), and Internet Exchange Points (IXPs) is available, it becomes possible to generate traffic in a more structured and reproducible way, directly linked to the infrastructure associated with each core node.
Our group has a topology-generation methodology and infrastructure-related information that describe how COs, DCs, and IXPs are distributed across the network. This provides the opportunity to design a traffic model in which COs, DCs, and IXPs drive traffic and exchange it among themselves for both intra-node and inter-node communication. The resulting model will produce inter-node traffic matrices while also preserving a portion of intra-node traffic, making it suitable for future simulation and planning studies.
The aim of this thesis is to develop and evaluate an infrastructure-aware traffic matrix generation model for optical core networks. The thesis will analyze how real infrastructure can be used to define the traffic-generation potential of each node and to construct reproducible traffic matrices that can later be used in realistic network simulations. Based on this analysis, the student will contribute to the creation of a reusable traffic-generation methodology that can be applied whenever the required infrastructure information is available.
Our group has a topology-generation methodology and infrastructure-related information that describe how COs, DCs, and IXPs are distributed across the network. This provides the opportunity to design a traffic model in which COs, DCs, and IXPs drive traffic and exchange it among themselves for both intra-node and inter-node communication. The resulting model will produce inter-node traffic matrices while also preserving a portion of intra-node traffic, making it suitable for future simulation and planning studies.
The aim of this thesis is to develop and evaluate an infrastructure-aware traffic matrix generation model for optical core networks. The thesis will analyze how real infrastructure can be used to define the traffic-generation potential of each node and to construct reproducible traffic matrices that can later be used in realistic network simulations. Based on this analysis, the student will contribute to the creation of a reusable traffic-generation methodology that can be applied whenever the required infrastructure information is available.
Target group:
Uni: BA oder MA für KT, ST, ME
Requirements:
Basic Python, basic graph theory
Tags:
Optical Core Networks, Traffic Matrix Generation, Infrastructure-Aware Modeling
Supervisor / Contact:
Christofer Vásquez
E-Mail: christofer.vasquez@unibw.de
E-Mail: christofer.vasquez@unibw.de