| The DO experiment is one of the two high energy physics experiments currently being conducted at Fermi National Accelerator Laboratory, in Batavia, IL, on what is currently the world's highest energy particle accelerator, the Tevatron. The experiment produces vast amounts of raw data of the order of several hundred terabytes. This data needs to be converted from the raw format that comes from the detector, ie. digitized data, to a format that is close to the physics, ie. data that can be subjected to analysis. This process, called reconstruction, is done according to constantly evolving and improving reconstruction algorithms. This process of conversion from raw detector data to analyzable data is done periodically to obtain data of very high quality, and it is done over the entire existing dataset to obtain consistent results. The scale of computation that is necessary to process such a large dataset is enormous and presents a challenge. This thesis discusses the computation problem that needs to be solved to achieve such a high level of scalability, the design of a computational model that will solve this problem, the issues that were encountered during implementation of the design, and steps taken to resolve these issues. |
