Speaker
Soon Yung Jun
(Fermi National Accelerator Lab. (US))
Description
Performance evaluation and analysis of large-scale computing
applications is essential for optimizing the use of resources. As detector
simulation is one of the most compute-intensive tasks and Geant4 is the
simulation toolkit most widely used in contemporary high energy
physics (HEP) experiments, it is important to monitor Geant4
through its development cycle for changes in computing performance and
to identify problems and opportunities for code improvements. All
development and public releases are being profiled with a set of
applications that utilize different input event samples, physics
lists, and detector configurations. Results from multiple
benchmarking runs are compared to previous public and development
reference releases to monitor CPU and memory usage. Observed changes are
evaluated and correlated with code modifications. Besides the full
summary of call stack data and memory footprint, a detailed call graph
analysis is available to Geant4 developers for further analysis. The
set of software tools used in the performance evaluation procedure,
both in sequential and multi-threaded modes, include FAST, IgProf and
Open$\mid$Speedshop. The scalability of the CPU time and memory
utilization of multi-threaded applications is evaluated by measuring
event throughput and memory usage as a function of the number of
threads for selected event samples. We will describe the procedure of
Geant4 computing performance profiling and benchmarking and present
recent results.
Authors
Andrea Dotti
(SLAC National Accelerator Laboratory (US))
Gunter Folger
(CERN)
Jim Kowalkowski
(Fermilab)
Krzysztof Genser
(Fermilab)
Dr
Marc Paterno
(Fermilab)
Soon Yung Jun
(Fermi National Accelerator Lab. (US))
Victor Daniel Elvira
(Fermi National Accelerator Lab. (US))
