Software-defined and programmable networking is a key enabling technology to support 5G and beyond networks in achieving their promises of increased scalability and flexibility at a lower cost.
Deep network programmability, that is the ability to program the network fabric both vertically (control and data plane) and horizontally (end to end), is expected to characterize the new generation of mobile networks (6G) towards supporting extreme performance requirements and service-specific operations. To this end, enriching next-generation mobile networks with data plane programming capabilities can bring significant benefits with regards to network monitoring and telemetry analysis. In-band network telemetry (INT) can help correlate observed application/network performance degradation with network bottlenecks, short-lived congestion events, routing misconfigurations, or highly utilized network nodes and links.
The objective of the post-doc appointment is to investigate in-band network telemetry (INT) solutions for evolving 6G systems with attention to 6G-optical integration. In this context, INT should support the collection of accurate end-to-end network state and QoS/QoE parameters, to support (proactive) network adaptation and near real time control. Thus, new approaches and technologies will be developed and demonstrated for collecting, intelligently aggregating (e.g., using including autoencoders and other ML-based techniques) and processing monitoring information originated by user equipment and terminals, programmable HW-accelerated network elements, edge computing resources, and applications. Envisioned solutions should provide high monitoring accuracy along with low and fixed transmission overhead and ensure scalability/performance.
This research will be carried out as part of longstanding collaboration with CIENA
, and it will result in joint demonstrators, proof-of-concepts and articles at suitable international conferences, such as Optical Fiber Communication Conference and Exhibition (OFC), ACM SIGCOMM, SuperComputing (SC), and in high impact journals. Testbeds available for this work are FABRIC
, CENI (Ciena environment for network innovation) and BRIDGES, which all have an international node located at the premises of the University of Amsterdam. Additional facilities will be provided as part of the UvA participation in the SLICES-RI
Furthermore, research synergies with high innovation potential will be explored with the EU HORIZON-JU-SNS-2022 project DESIRE6G (DEep programmability and Secure distributed Intelligence for Real-time End-to-end 6G networks) spearheaded by the MNS team.
You will be embedded in the MNS group. The group focuses its researches on the fundamental architectural problems that arise from the interconnection of systems and of data flows. We look at the emerging architectures that can support the operations of the future Internet. More information can be found at the MNS website
.Tasks and responsibilities
- focus on the development of a data-plane telemetry system for pervasive monitoring and intelligent data aggregation for 6G systems;
- benchmark the performance or validate the applicability of proposed approaches;
- jointly work in the FABRIC research community on large-scale demonstrators;
- investigate synergies with the BRIDGES and the SLICES-RI initiatives;
- become active in the DESIRE6G research community and collaborate with other institutes and/or companies that are part of the project;
- publish and present work regularly at international conferences, workshops, and journals;
- support relevant educational activities and provide graduation opportunities to bachelor and master students on topics related to the research.