Porting WarpX to GPU-accelerated platforms

Myers, Andrew C.; Almgren, Ann S.; Amorim, L. D.; Bell, John B.; Fedeli, Luca; Ge, Lixin; Gott, Kevin; Grote, D.P.; Hogan, Mark; Huebl, Axel; Jambunathan, Revathi; Lehe, Rémi; Ng, Cho-Kuen; Rowan, Michael E.; Shapoval, Olga V.; Thévenet, Maxence; Vay, Jean‐Luc; Vincenti, Henri; Yang, Erchan; Zaïm, N.; Zhang, Weiqun; Zhao, Yinjian; Zoni, Edoardo

doi:10.1016/j.parco.2021.102833

Cited by 40 publications

(25 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a case study we consider the tuning of two lenses separating three consecutive plasma stages. A GPU-optimized version of WarpX [26] was used to simulate each design in a 192 × 192 × 7680 grid with 50,000 particles. This resulted in a typical simulation runtime of 25 minutes on 2 nodes of the OLCF Summit system (each node contains 6 NVIDIA Voltas).…”

Section: Designing Plasma-based Accelerator Stagesmentioning

confidence: 99%

libEnsemble: A Library to Coordinate the Concurrent Evaluation of Dynamic Ensembles of Calculations

Hudson,

Larson,

Navarro

et al. 2021

Preprint

View full text Add to dashboard Cite

Almost all applications stop scaling at some point; those that don't are seldom performant when considering time to solution on anything but aspirational/unicorn resources. Recognizing these tradeoffs as well as greater user functionality in a near-term exascale computing era, we present libEnsemble, a library aimed at particular scalability-and capability-stretching uses. libEnsemble enables running concurrent instances of an application in dynamically allocated ensembles through an extensible Python library. We highlight the structure, execution, and capabilities of the library on leading pre-exascale environments as well as advanced capabilities for exascale environments and beyond.

show abstract

Section: Designing Plasma-based Accelerator Stagesmentioning

confidence: 99%

libEnsemble: A Library to Coordinate the Concurrent Evaluation of Dynamic Ensembles of Calculations

Hudson,

Larson,

Navarro

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Because of their flexibility and versatility, PIC schemes are employed in many production level plasma simulation and particle accelerator codes such as TRISTAN-MP [5,6], ORB5 [7], XGC [8], OSIRIS [9], IMPACT-T [10], OPAL [11] and Warp-X [12], to name a few. Some of these production codes have already begun their journey towards performance portability as evidenced in [13,14]. It is expected that a lot more will also do so sometime soon, in order to benefit from the high performance of existing and future advanced computing architectures.…”

Section: Introductionmentioning

confidence: 99%

“…• So far, portable exascale PIC studies have mostly been conducted for PIC schemes designed for electromagnetic plasma models [14,16]. To the best of our knowledge, this is the first study which considers the performance of a PIC scheme for an electrostatic plasma model in the portable exascale context.…”

Section: Introductionmentioning

confidence: 99%

ALPINE: A set of performance portable plasma physics particle-in-cell mini-apps for exascale computing

Muralikrishnan¹,

Frey²,

Vinciguerra³

et al. 2022

Preprint

View full text Add to dashboard Cite

Alpine consists of a set of mini-apps that makes use of exascale computing capabilities to numerically solve some classical problems in plasma physics. It is based on IPPL (Independent Parallel Particle Layer), a framework that is designed around performance portable and dimension independent particles and fields. In this work, IPPL is used to implement a particle-in-cell scheme.The article describes in detail the following mini-apps: weak and strong Landau damping, bump-on-tail and two-stream instabilities, and the dynamics of an electron bunch in a charge-neutral Penning trap. We benchmark the simulations with varying parameters such as grid resolutions (512 3 to 2048 3 ) and number of simulation particles (10 9 to 10 11 ). We show strong and weak scaling and analyze the performance of different components on several preexascale architectures such as Piz-Daint, Cori, Summit and Perlmutter.

show abstract

“…open-source nor optimized for large-scale GPU-based HPC machines. This paper presents PICSAR-QED [42], a module part of the PICSAR library [43], which has been coupled with the WarpX [44,45] PIC code in order to simulate sf-QED processes relevant for extreme plasma physics scenarios. WarpX is an opensource code developed within the framework of the Exascale Computing Project [46] and designed to provide scalable performance on upcoming exascale supercomputers.…”

Section: Introductionmentioning

confidence: 99%

PICSAR-QED: a Monte Carlo module to simulate Strong-Field Quantum Electrodynamics in Particle-In-Cell codes for exascale architectures

Fedeli,

Zaïm,

Sainte-Marie

et al. 2021

Preprint

View full text Add to dashboard Cite

Physical scenarios where the electromagnetic fields are so strong that Quantum ElectroDynamics (QED) plays a substantial role are one of the frontiers of contemporary plasma physics research. Investigating those scenarios requires state-ofthe-art Particle-In-Cell (PIC) codes able to run on top high-performance computing machines and, at the same time, able to simulate strong-field QED processes. This work presents the PICSAR-QED library, an open-source, portable implementation of a Monte Carlo module designed to provide modern PIC codes with the capability to simulate such processes, and optimized for high-performance computing. Detailed

show abstract

Porting WarpX to GPU-accelerated platforms

Cited by 40 publications

References 21 publications

libEnsemble: A Library to Coordinate the Concurrent Evaluation of Dynamic Ensembles of Calculations

libEnsemble: A Library to Coordinate the Concurrent Evaluation of Dynamic Ensembles of Calculations

ALPINE: A set of performance portable plasma physics particle-in-cell mini-apps for exascale computing

PICSAR-QED: a Monte Carlo module to simulate Strong-Field Quantum Electrodynamics in Particle-In-Cell codes for exascale architectures

Contact Info

Product

Resources

About