Watering the Seed for Speed

This project focused on optimising track seeding in FastTrackFinder (FTF) on a GPU using Nvidia tools to increase speed without compromising efficiency. The processing time for track seeding within FTF was reduced from 153.45 ± 4.09 ms to 115.97 ± 2.75 ms, while maintaining the same level of accuracy. In addition, a study of multi-processing configurations revealed that performance is primarily influenced by the number of concurrent events rather than the configuration, allowing for the selection of configurations that best suit the needs of other algorithms.