The determination of membrane permeability is a necessary step in the development of any new drug, as an inability to cross the cellular membrane will severely limit its efficacy. Currently, this permeability must be assayed experimentally, for potentially thousands of candidates. Computational molecular dynamics simulations represent a way to estimate permeabilities without going through costly synthesis, provided they are sufficiently ac- curate. Here, we propose to test their accuracy by, for the first time, using a membrane model matching the one used experimentally. The simulations require running multiple systems simultaneously, each differing slightly in the placement of the drug. Not only are these simulations best suited for Blue Waters, but once tested, scaling our methods up to screen hundreds or thousands of candidates rapidly can be accomplished nowhere else.