Complex Systems are composed of large, interacting entities which taken together, exhibit a macroscopic behavior which is not predictable by examination of the individual entities. The Complex Systems Program seeks to understand the fundamental science of these systems and develop rigorous descriptions (analytic, statistical, or semantic) that enable prediction and control of their behavior. Initially focused on the Internet and Grid Computing, this Program will facilitate predictability and reliability in these areas. The forthcoming report "Study of Proposed Internet Congestion Control Mechanisms" represents a state-of-the-art experiment examining Internet Congestion Control. Equally as important, the methodology used in the experiment provides the clearest example of experimental design applied to this domain and is a major contribution to the field. In the future, we hope to apply these complex systems capabilities to other domains such as biotechnology, nanotechnology, semiconductors, and engineering in support of the other laboratories at NIST.