Co-design of a Particle-in-Cell Plasma Simulation Code for Intel Xeon Phi: A First Look at Knights Landing

Surmin, Igor A.; Bastrakov, Sergei; Matveev, Zakhar A.; Efimenko, Evgeny; Gonoskov, Arkady; Meyerov, Iosif B.

doi:10.1007/978-3-319-49956-7_25

Cited by 17 publications

(16 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PIConGPU [18], UPIC [8], ORB5 [9], PI-CADOR [19], PICSAR [6]. The L6D space-filling curve we propose can be seen as a technique leading to a similar organization of the particles: particles are kept together in memory also in a block-like fashion, cf.…”

Section: Related Workmentioning

confidence: 99%

Efficient data layouts for a three-dimensional electrostatic Particle-in-Cell code

Barsamian

Hirstoaga

Violard

2018

Journal of Computational Science

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Efficient data layouts for a three-dimensional electrostatic Particle-in-Cell code

Barsamian

Hirstoaga

Violard

2018

Journal of Computational Science

View full text Add to dashboard Cite

“…QED PIC simulations are very computationally intensive and therefore are performed on supercomputers using highly optimized parallel software [10][11][12][13][14][15][16][17][18]. The problem of efficient implementation of the PIC method for parallel machines is quite well studied [19][20][21][22][23][24][25][26][27]. Fortunately, the method has a large potential for parallelization due to the local nature of the interactions.…”

Section: Introductionmentioning

confidence: 99%

Exploiting Parallelism on Shared Memory in the QED Particle-in-Cell Code PICADOR with Greedy Load Balancing

Meyerov

Панов

Bastrakov

et al. 2020

Parallel Processing and Applied Mathematics

Self Cite

View full text Add to dashboard Cite

State-of-the-art numerical simulations of laser plasma by means of the Particle-in-Cell method are often extremely computationally intensive. Therefore there is a growing need for development of approaches for efficient utilization of resources of modern supercomputers. In this paper, we address the problem of a substantially non-uniform and dynamically varying distribution of macroparticles in a computational area in simulating quantum electrodynamic (QED) cascades. We propose and evaluate a load balancing scheme for shared memory systems, which allows subdividing individual cells of the computational domain into work portions with subsequent dynamic distribution of these portions between OpenMP threads. Computational experiments on 1D, 2D, and 3D QED simulations show that the proposed scheme outperforms the previously developed standard and custom schemes in the PICADOR code by 2.1 to 10 times when employing several Intel Cascade Lake CPUs.

show abstract

“…A first approach to mitigate randomness in PIC codes is to use the standard domain decomposition on domains so small that they can fit in the cache of the system. This method is commonly used in recent implementations and the small domains are often referred to as patches [2] or tiles [6,7] (from now on the term patch is used as the generic denomination). It exposes a very high level of parallelism and mitigates memory access randomness since particles of each patch all access the same grid region which is limited by the patch extension.…”

Section: Introductionmentioning

confidence: 99%

Adaptive SIMD optimizations in particle-in-cell codes with fine-grain particle sorting

Beck

Dérouillat

Lobet

et al. 2019

Computer Physics Communications

View full text Add to dashboard Cite

Particle-In-Cell (PIC) codes are broadly applied to the kinetic simulation of plasmas, from laser-matter interaction to astrophysics. Their heavy simulation cost can be mitigated by using the Single Instruction Multiple Data (SIMD) capibility, or vectorization, now available on most architectures. This article details and discusses the vectorization strategy developed in the code Smilei which takes advantage from an efficient, systematic, cell-based sorting of the particles. The PIC operators on particles (projection, push, deposition) have been optimized to benefit from large SIMD vectors on both recent and older architectures. The efficiency of these vectorized operations increases with the number of particles per cell (PPC), typically speeding up three-dimensional simulations by a factor 2 with 256 PPC. Although this implementation shows acceleration from as few as 8 PPC, it can be slower than the scalar version in domains containing fewer PPC as usually observed in vectorization attempts. This issue is overcome with an adaptive algorithm which switches locally between scalar (for few PPC) and vectorized operators (otherwise). The newly implemented methods are benchmarked on three different, large-scale simulations considering configurations frequently studied with PIC codes.

show abstract

Co-design of a Particle-in-Cell Plasma Simulation Code for Intel Xeon Phi: A First Look at Knights Landing

Cited by 17 publications

References 22 publications

Efficient data layouts for a three-dimensional electrostatic Particle-in-Cell code

Efficient data layouts for a three-dimensional electrostatic Particle-in-Cell code

Exploiting Parallelism on Shared Memory in the QED Particle-in-Cell Code PICADOR with Greedy Load Balancing

Adaptive SIMD optimizations in particle-in-cell codes with fine-grain particle sorting

Contact Info

Product

Resources

About