Computing for better accelerators

Panagiotis Spentzouris

Panagiotis Spentzouris, principal investigator of the ComPASS project and head of the Accelerator and Detector Simulation Department, wrote this column.

The path for scientific discovery at the three frontiers of high-energy physics is paved with advances in theory, experiment and simulation. In accelerator science, numerical modeling and simulation are essential for the development of new particle acceleration concepts and technologies and for the design, optimization and successful operation of accelerators.

Because of the complexity of accelerator components and the many physics processes involved in describing the behavior of particle beams as they travel through these components, high-fidelity numerical models require massive computations on powerful supercomputers. These models allow accelerator scientists to achieve higher accuracy, shorten the turnaround time in designing and optimizing accelerators, and reduce cost by lowering the number of trial-and-error cycles needed for producing accelerator component prototypes.

The Community Project for Accelerator Science and Simulation (ComPASS) is a collaboration of about 20 senior scientists – computational accelerator scientists, computer scientists and mathematicians from national laboratories, universities and private companies. Its mission is to develop simulation tools that can effectively make use of extreme-scale computing resources. Recently, ComPASS was awarded $1.8 million per year for three years under the third installment of the DOE program for Scientific Discovery through Advanced Computing (SciDAC3) through a partnership with DOE’s High Energy Physics and Advanced Scientific Computing Research programs.

Under SciDAC3, ComPASS will deploy applications that will enable accelerator science advances at the Intensity and Energy Frontiers. Collaborating with other accelerator scientists, the ComPASS researchers use simulations to:

  • develop new technologies and techniques that maximize particle acceleration in the shortest possible distance,
  • investigate the potential of new techniques, technologies and materials for high-gradient acceleration,
  • optimize the design of Intensity Frontier accelerators such as Project X, and
  • explore the options for the next generation of Energy Frontier machines.

We face many challenges as we push to advance accelerator technology using cutting-edge computing technology, which are both evolving at the same time. With the support of the accelerator community and the DOE programs, the problem-solving minds of the ComPASS collaboration will meet these challenges.