The rapid information flow from the experiments being constructed at the Large Hadron Collider (LHC) in the European Laboratory for High Energy Physics (CERN), Geneva presents a major computational challenge. In this ITR proposal, the UW-Madison team proposes to design and build a prototype Rapid-Response Adaptive Computing Environment (RACE) that integrates their local resources with the globally distributed computing grid, to cater to mission-critical, time-bound needs of the CMS experiment. These RACEs will have resource management features as well as knowledge management features. The UW team proposes to develop RACE technology by focusing on CMS trigger system data validation and data forensics software.
This work can empower individual physicists and small groups enabling efficient, reliable, tractable and timely extraction of LHC physics. In addition, they propose to help adapt the RACE set-up to the needs of the UW Medical Physics researchers who expect to exploit commodity computers for providing burst computing for their online radiation therapy applications. The team anticipates that the software infrastructure that they develop for knowledge management can be applied to other domains by developing appropriate ontologies and descriptions using the same overall principles. The graduate research assistants working on this project will benefit from applying their research ideas to a concrete real-world application. The cross-fertilization of ideas from their interdisciplinary team should enhance their ability to come up with innovative ideas when they move on to work in the industry.
