Ryohei Sugiura ; Yutaka Ushiroda

2020
The University of Tokyo Bunkyo

Abstract: The Belle II experiment is a B-factory experiment hosted by KEK in Tsukuba, Japan. Belle II experiment is the successor of the Belle experiment, which observed CP violation in the B-meson system. Belle II experiment observes decay products of the B-meson pairs created by the SuperKEKB accelerator by the Belle II detector to find new physics beyond the standard model. The luminosity of e$^-$ and e$^+$ collision of SuperKEKB accelerator is upgraded to 40 times higher than that of KEKB accelerator in Belle experiment to obtain abundant statistics, thereby the trigger rate for Belle II experiment is expected to increase up to 30kHz. Therefore, in the data acquisition system we need to handle large data throughput from the whole Belle II detector under a sufficiently small dead time. To supervise and evaluate dead time in data acquisition, a monitoring system for dead time in the data acquisition system is implemented in this thesis. The performance of the data acquisition system in 2019 physics run is evaluated using the monitoring system, and it is confirmed that we can obtain data under 0.1\% dead time with a 5kHz trigger rate. In addition to the dead time monitoring, possible bottlenecks of data flow in the data acquisition system are evaluated. If the data throughput rate exceeds the bandwidth of the potential bottlenecks, the large dead time arises due to the backpressure. Therefore, the data rate under high luminosity must be below the potential bottlenecks. In this thesis, the data size under high luminosity is evaluated, and the effect of planned data acquisition system upgrades on those potential bottlenecks is discussed.

Note: Presented on 28 01 2020
Note: MSc