I'm looking to run scheduled accumulative builds using large datasets and was wondering if there is a way to verify that this process is not, in fact, rebuilding the entire cube on each scheduled run.  Is there a log file or the like that details the build process?  Thanks!