The KDI team in the Center for Space Environment Modeling is a multi-disciplinary group collaborating to develop advanced computing and communications technology. Using this technology, we are working to produce the first prototype implementation of a first-principles, physics-based Comprehensive Space Environment Model. This objective is motivated by the needs expressed in the National Space Weather Program Strategic Plan (FCM-P30-1995, Office of the Federal Coordinator for Meteorological Services and Supporting Research, Silver Spring, MD. 1995). This document notes the enormous value of our nation's technology deployed in space and on the ground which, as it becomes more sophisticated, also becomes more vulnerable to various dynamic phenomena which occur in the near-Earth space environment. As a consequence, a national goal has been set to produce a physics-based model of the space environment, capable of providing accurate predictions of the environment in order to enable the operators of technologies to undertake mitigating practices to protect their assets from 'space weather' storms. Accomplishing this requires the combined efforts of a wide range of experts, and the solution to many technical challenges which are at the core of the KDI solicitation. Further, the technical challenges which we are solving are relevant to a wide range of applications beyond the specific modeling effort which we are undertaking. Therefore in response to the goals of the KDI solicitation and to the needs of the National Space Weather Program, our purpose is to develop a seamless flow of knowledge regarding space weather events and hazards based on real time observations which are incorporated into validated models to produce results for use by researchers, space weather forecasters, students and the public. This requires that we develop high performance computing techniques which address the critical processes of the Sun-Earth system which occur over 9 orders of magnitude in scale, that we utilize the dynamic interplay between observation and numerical models to achieve the most accurate results, that we develop algorithm efficiencies which enable the models to run faster than real time, and that we develop, deploy, utilize and evaluate an electronic knowledge network, or collaboratory, to support the globally distributed team which will develop, test, validate and utilize the output of the Comprehensive Space Environment Model that will be produced by this effort.