Researchers have turned to XSEDE to realize their vision of a world in which an energy model can attain cost savings, security, and sustainability in buildings. A team of researchers at Oak Ridge National Laboratory are leading a project known as Autotune, working to develop a methodology which places the bargain of energy efficiency within reach for more commercial and residential buildings. Compute time on XSEDE-allocated Nautilus supercomputer is helping the team create a more accurate energy model than exists today. This advanced analytical and optimization methodology leverages terabytes of HPC-generated simulation data and data mining with multiple-machine learning algorithms for quickly calibrating a building energy model to measured (utility or sensor) data. The Autotune project has clocked approximately 300,000 compute hours on Nautilus -- completing about 1.6 million parametric simulations of the eventual 8 million and generating more than 50 terabytes of data of the planned 270 terabytes, so far. To read further, please visit http://www.nics.tennessee.edu/autotune.