Network virtualization enables flexible placement, migration, and execution of virtual networks and machines on physical hardware. This results in an NPhard optimization problem called virtual network embedding (VNE). Ensuring hardware and (non-)functional constraints while finding an optimal solution for a wide range of scenarios is a challenging task in these highly dynamic environments. To develop and evaluate algorithms tailored for various environments, we present a model-driven approach to specify and solve dynamic VNE problems by using a high-level specification to declaratively specify the search space and constraints, incremental model transformation to prune the search space, and low-level ILP techniques to find an optimal solution in the pruned search space. This high-level specification is used to generate an executable program for solving dynamic VNE problems. Furthermore, we show in an evaluation that this generated program can solve a typical dynamic VNE problem in less time than a hand-crafted ILP-based program.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.