Computational discovery of cis-regulatory motifs has become one of the more challenging problems in bioinformatics. In recent years, over 150 methods have been proposed as solutions, however, it remains difficult to characterize the advantages and disadvantages of these approaches because of the wide variability of approaches and datasets. Although biologists desire a set of parameters and a program most appropriate for cis-regulatory discovery in their domain of interest, compiling such a list is a great computational challenge. First, a discovery pipeline for 150+ methods must be automated and then each dataset of interest must used to grade the methods. Automation is challenging because these programs are intended to be used over a small set of sites and consequently have many manual steps intended to help the user in fine-tuning the program to specific problems or organisms. If a program is fine-tuned to parameters other than those used in the original paper, it is not guaranteed to have the same sensitivity and specificity. Consequently, there are few methods that rank motif discovery tools. This paper proposes a parallel framework for the automation and evaluation of cis-regulatory motif discovery tools. This evaluation platform can both run and benchmark motif discovery tools over a wide range of parameters and is the first method to consider both multiple binding locations within a regulatory region and regulatory regions of orthologous genes. Because of the large amount of tests required, we implemented this platform on a computing cluster to increase performance.