Pebbles mechanics simulation speedup

Pebble bed reactors contain large numbers of spherical fuel elements arranged randomly. Determining the motion and location of these fuel elements is useful for calculating certain operating parameters of pebble bed reactors. These simulations involve determining the entire core motion as hundreds of thousands of pebbles are recirculated over months to years of physical time. Single processor algorithms are insufficient for this simulation because they would take decades to centuries of wall-clock time. This paper describes the process of parallelizing and speeding up the PEBBLES mechanics simulation code. Shared memory programming with the Open Multiprocessing Application Programming Interface and distributed memory programming with the Message Passing Interface are used simultaneously in this process. A new shared memory lock-less linear time collision detection algorithm is described, a method that allows faster detection of pebbles in contact than generic methods. This paper also provides an overview of the computationally expensive portions of computer codes and the different areas to be optimized. These improvements to PEBBLES combine to make full simulations of recirculations in German AVR-sized reactors possible in months of wall clock time.