The number of Distributed Green Cloud Datacenters (DGCDs) is globally increasing. Such DGCDs deploy different types of renewable sources to generate clean energy and save money. They are located in different regions depending on the availability of renewable energy sources, costs of bandwidth and grid electricity prices. This paper focuses on applications within them, which are sensitive to delay and looks into a way to schedule different applications with respect to time constraint. The paper uses the Firefly algorithm, bat algorithm and simulated annealing-bat algorithms as optimization techniques to minimize total operational cost of such DGCDs. These algorithms have been compared through data-driven experiments carried out in this study. Of particular note is the Firefly algorithm's superior performance when compared with others.