We investigate the problem of optimally placing virtual network functions in 5G-based virtualized infrastructures according to a green paradigm that pursues energy-efficiency. This optimization problem can be modelled as an articulated 0-1 Linear Program based on a flow model. Since the problem can prove hard to be solved by a state-of-the-art optimization software, even for instances of moderate size, we propose a new fast matheuristic for its solution. Preliminary computational tests on a set of realistic instances return encouraging results, showing that our algorithm can find better solutions in considerably less time than a state-of-the-art solver.
CITATION STYLE
Bauschert, T., D’Andreagiovanni, F., Kassler, A., & Wang, C. (2019). A matheuristic for green and robust 5G virtual network function placement. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11454 LNCS, pp. 430–438). Springer Verlag. https://doi.org/10.1007/978-3-030-16692-2_29
Mendeley helps you to discover research relevant for your work.