Communication networks are currently undergoing substantial changes, which are partly driven by new technologies (e.g. SDN, NFV), and partly by new requirements, such as reduced latency and operational costs, or increased reliability. It is a challenge to provide robust and performant control planes and connectivity in such highly flexible and demanding networking environments, since typically resources are shared. Also, network debugging and diagnostics need to cope with these new demands.
Managing and operating such networks becomes increasingly challenging and calls for new approaches that can tackle the inherently increasing complexity introduced into flexible software-defined and virtualized communication networks of the future. In this sense, communication networks should increasingly operate autonomously without manual intervention. A more flexible usage of costly network infrastructures could help to radically cut the service provisioning time, while keeping the total cost of ownership low. This also comprises a flexible usage of network infrastructure across different tenants, with data centers of different sizes and at different locations being part of this infrastructure. Moreover, suitable distributed runtime scheduling algorithms are required in order to utilize and share network resources efficiently and to fulfill highly demanding requirements from certain network slices, e.g., ultra-reliable low latency communication in case of industrial networks. It needs to be investigated, to what extent artificial intelligence and machine learning can be applied in this context, and we should consider potential synergies at least among the following aspects of automation in communication systems:
- Deployment and dynamic adjustment of (virtualized) network and upper-layer services, including demand-driven relocation of functionalities and services.
- Robust and performant control planes in highly dynamic and autonomous communication systems that share common networking resource potentially also with the data plane.
- Network debugging and diagnostics (e.g., automated detection of routing failures or DDoS attacks).
Properly addressing these topics requires a multitude of expertise topics, covering an area that we can describe as new Elastic, robust and performant connectivity, which will have to rely on distributed runtime resource scheduling, coupled with Machine Learning in network control. This requires researchers to extend their action scope and to bridge concepts and challenges across these different disciplines. Accordingly, we seek for consensual insights from experts in “classical” networking, distributed systems and machine learning for networks.
During this short Dagstuhl Seminar, we therefore plan to bring together leading experts from different backgrounds, to explore potential synergies as mechanisms to address the upcoming challenges on the automation of communication networks. Each day we plan to have three plenary discussions, followed by separate working groups, which will progress on the seminar challenges. As an outcome of the seminar, besides new exciting collaborations, we plan to develop a white paper summarizing the state of the art and identify the most important and challenging directions for future research in this area.
- Communication Networks