Network virtualization provides an effective means of supporting multiple client infrastructures over a physical substrate. Now over the years, many different provisioning algorithms have been developed for network virtualization services, most recently driven by the growth in new cloud-computing paradigms. However, as user demands continue to evolve, there is growing need to schedule virtual network requests at future time instants. This is a very complex problem given the high dimensionalities involved. Along these lines this paper solves one of the first optimization formulations for virtual network scheduling and compares the findings to some advanced heuristics.