The ATAP Division’s Advanced Modeling Program (AMP) develops and uses cutting-edge computing techniques and software, working with national and international collaborators to achieve fast and accurate simulations of particle beams and accelerators, plasmas and fusion devices, and more. Our approach draws on deep collaboration among physicists, applied mathematicians, and computer scientists—a synergy of Berkeley Lab’s strengths. The results benefit many fields of the DOE Office of Science research portfolio.

We lead two multi-laboratory Scientific Discovery through Advanced Computing (SciDAC) collaborations: the DOE Office of High Energy Physics-supported Collaboration for Advanced Modeling of Particle Accelerators (CAMPA) and the DOE Office of Fusion Energy Sciences-supported Kinetic IFE Simulations at Multiscale with Exascale Technologies collaboration (KISMET).

Simulation result slices open the collision between positron and electron beams

AMP develops and applies computational tools to address the most challenging problems in beam physics, high-fidelity modeling of beam transport, and advanced accelerator design and concepts. Our long-term goal is to use computational tools to virtually prototype, design, and optimize entire accelerators self-consistently and fast enough for real-time feedback and accelerator auto-tuning.

Magnetic Vortex Acceleration simulation shows a particle beam emerging from a tubular plasma

AMP’s tools and methods are used to model fundamental plasma physics, laser-plasma interaction, high-energy-density science, high-field physics, laboratory and astrophysical plasmas, and fusion physics. Including the complex interactions of charged particles, fields, and structures from first principles and reduced models enables the modeling of complex plasma and fusion devices.

Simulation performed with the WarpX code shows surfaces of equal potential in a laser plasma accelerator

Computing is at the heart of AMP’s activities, spanning the development and analysis of novel algorithms, their implementation in AMP’s codes, and optimization on the fastest and largest supercomputers; the exploration and use of AI/ML; the investigation of algorithms for quantum computers; and the development of data standards for software interoperability and the establishment of software ecosystems.

Beam, Plasma, & Accelerator Simulation Toolkit (BLAST) logo

BLAST is the backbone of a coherent ecosystem of advanced codes, libraries, interfaces, data standards, and AI/ML integrations. It is developed and maintained as an open-source repository by and for an international community. AMP is deeply involved in all aspects of BLAST and leads, co-leads, or co-develops the various components of the toolkit.

Simulation shows a laser pulse impacting a target and causing emission and acceleration of a high-charge electron bunch

The “exascale” era of computing opens up revolutionary opportunities for optimizing designs and discovering and exploring new ideas. AMP focuses on making the most of this game-changing capability in an activity rooted in DOE’s Exascale Computing Project. A prime example is the code WarpX, which was honored with the Association for Computing Machinery’s Gordon Bell Prize.

Logo of the SciDAC program

The Advanced Modeling Program leads two multi-laboratory Scientific Discovery through Advanced Computing (SciDAC) collaborations: the DOE Office of High Energy Physics-supported Collaboration for Advanced Modeling of Particle Accelerators (CAMPA) and the DOE Office of Fusion Energy Sciences-supported Kinetic IFE Simulations at Multiscale with Exascale Technologies collaboration (KISMET).

Publications