Ongoing research deals with the challenges of resource managing distributed cloud infrastructures embedded in the telecommunications infrastructure. The main contribution lies in modelling the infrastructure’s resources and the applications that reside within it, as a means to formulate a optimisation problem to minimise the operational cost of such an infrastructure. The main objective of the research effort is to develop a scalable and distributed algorithm for continuous operational cost minimisation that ensures that the resident applications performance requirements are met. To successfully do so the algorithm needs to take into accommodate a set of highly dynamic inputs as well as the system’s internal state. Tangent with this work we are also exploring control theoretical methods for dynamically allocating a shared network resource between multiple traffic flows with heterogenous performance objectives.
Contact: William Tärneberg