Networks are rapidly evolving from a facility for mere data transport to an infrastructure for the execution of services. Examples include so-called “edge computing” in systems like 5G or provisioning load balancers directly in a switch in a data centre, saving dozens to hundreds of servers. In this talk, we will look at some of the challenges when we want to execute complex services in such an infrastructure, catering to large user populations, resulting in challenging optimisation and operations problems. A particularly interesting example are services that use or train machine-learning models: their requirements are significantly different from ordinary services and provide a research challenge for the foreseeable future.
The event starts at 5:00 p.m. in the lecture hall in House L (Campus II).
