Summary: | IceCube is a one cubic kilometer international high-energy neutrino observatory at the South Pole. The IceCube experiment has two main components: InIce and IceTop. InIce located approximately 1500 meters below the surface of ice would deploy DOMs over 80 strings with 60 DOMs on each string. IceTop, located on the surface comprises an array of DOMs and would deploy 320 DOMs over 80 stations. ☐ The data acquisition component of this experiment, known as the DAQ, is a collection of hardware and software components. There are several challenges involved in this experiment because of the real-time nature of the data being generated, large volumes of data to be handled without any losses, limited resources like power, network issues and extreme weather conditions at the site of deployment. The work done in this thesis is specifically in context of IceTop. ☐ A part of this work focuses on improving a specific component of IceCube DAQ called the Splicer. The Splicer is a software module responsible for merging several input sources and producing a single ordered stream. It has been used extensively in the whole IceCube experiment. We have analyzed performance of the Splicer using several methodologies. Out of the several techniques experimented, the most effective was to reduce Garbage Collection and Object Cycling. ☐ A second component that was designed and evaluated as part of this thesis was Monitoring System. The evaluation was done under different loads and various refresh rates to ensure the correctness and stability of the monitoring system. Seckel, David Swany, Martin M.S. University of Delaware, Department of Computer and Information Sciences
|