Grid Computing for LHC and Methods for W Boson Mass Measurement at CMS

Abstract

In the Standard Model (SM) of particle physics, the masses of the Higgs boson, of the W boson and of the top quark are correlated via electroweak loop corrections. The experiments at the Large Hadron Collider (LHC) focus, amongst other goals, on checking the consistency of the SM and searching for 'new physics'. This thesis studies the measurement of the W boson mass at the Compact Muon Solenoid (CMS) experiment. At the LHC, a very large amount of data will be recorded because of the unprecedented center of mass energy and luminosity. As the computing resources for efficient data storage and processing exceed capabilities of a centralized computing approach, the LHC experiments have embraced grid computing. By using a tiered structure, different tasks are distributed to computing centers worldwide; still, the user has an easy access to all the resources. The feasibility study of the W boson measurement presented in this thesis requires a large number of proton proton collisions to be simulated. As the computing demands of the simulation exceed the local resources, the analysis has relied on the use of computing resources on the grid. In this context, the local computing cluster was integrated into the grid and administrated. The LHC experiments will record a much higher number of W boson and Z boson events than previous experiments. New methods, which use this higher statistics, can be employed for the W boson measurement. Two of these are the 'scaling method' and the 'morphing method'. The scaling method compares observable distributions from W and Z boson events, while the morphing method uses an analytical transformation of Z boson events for modelling W boson events. The methods are investigated in their statistical resolution and in their systematic uncertainty for the effects of the early CMS detector in this thesis.