IDLab-NetModel has 25+ years of research experience in the domains of optical networking, routing and recovery mechanisms, software-defined networking, graph modelling, optimization and green networking. These activities are ranging from fundamental to applied research and are conducted within the framework of European projects, national projects and bi-lateral research with industrial partners. This research has resulted in hundreds of publications in scientific top-journals and renowned international conferences.
This involves fundamental research to address complex graph problems in network science. We tackle challenges related to graph modeling, routing, and network optimization amongst others. Our expertise extends to both discrete and continuous optimization and simulation, encompassing online and offline settings. Additionally, we delve into the design and analysis of related data structures. By combining insights from mathematics and computer science (graph theory, linear programming, AI), we strive to create next-generation algorithms that find optimal solutions amidst vast possibilities, analyze their performance and prove their correctness.
This research bridges the gap between software and hardware. IDLab-Netmodel performs research in performance profiling, resource allocation and orchestration algorithms on top of heterogeneous hardware involving accelerator options. IDLab-Netmodel has an excellent track record in this domain and can build further on extensive expertise in SDN and NFV control mechanisms built in the context of national and international research projects.
IDLab-Netmodel designs and examines novel network architectures and routing mechanisms for Photonic Integrated Chips or multi-chip(let) processing architectures including optical Network-on-Wafer (NoW) systems. We are also investigating interconnects for High Performance Computing (HPC), a key platform for the training of massive machine learning models or huge simulations. IDLab-Netmodel is researching network interconnect architectures and resource management strategies for HPC systems, to enable efficient execution of multiple massive workloads avoiding bottlenecks in the network.
