Research output: Contribution to journal › Article › peer-review
About improving efficiency of the P3M algorithms when computing the inter-particle forces in beam dynamics. / Kozynchenko, Alexander I.; Kozynchenko, Sergey A.
In: Computer Physics Communications, Vol. 212, 01.03.2017, p. 47-54.Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - About improving efficiency of the P3M algorithms when computing the inter-particle forces in beam dynamics
AU - Kozynchenko, Alexander I.
AU - Kozynchenko, Sergey A.
N1 - Publisher Copyright: © 2016 Elsevier B.V. Copyright: Copyright 2016 Elsevier B.V., All rights reserved.
PY - 2017/3/1
Y1 - 2017/3/1
N2 - In the paper, a problem of improving efficiency of the particle–particle– particle-mesh (P3M) algorithm in computing the inter-particle electrostatic forces is considered. The particle-mesh (PM) part of the algorithm is modified in such a way that the space field equation is solved by the direct method of summation of potentials over the ensemble of particles lying not too close to a reference particle. For this purpose, a specific matrix “pattern” is introduced to describe the spatial field distribution of a single point charge, so the “pattern” contains pre-calculated potential values. This approach allows to reduce a set of arithmetic operations performed at the innermost of nested loops down to an addition and assignment operators and, therefore, to decrease the running time substantially. The simulation model developed in C++ substantiates this view, showing the descent accuracy acceptable in particle beam calculations together with the improved speed performance.
AB - In the paper, a problem of improving efficiency of the particle–particle– particle-mesh (P3M) algorithm in computing the inter-particle electrostatic forces is considered. The particle-mesh (PM) part of the algorithm is modified in such a way that the space field equation is solved by the direct method of summation of potentials over the ensemble of particles lying not too close to a reference particle. For this purpose, a specific matrix “pattern” is introduced to describe the spatial field distribution of a single point charge, so the “pattern” contains pre-calculated potential values. This approach allows to reduce a set of arithmetic operations performed at the innermost of nested loops down to an addition and assignment operators and, therefore, to decrease the running time substantially. The simulation model developed in C++ substantiates this view, showing the descent accuracy acceptable in particle beam calculations together with the improved speed performance.
KW - Charged particle beam
KW - Inter-particle forces
KW - N-body
KW - PM algorithms
KW - Poisson's equation
UR - http://www.scopus.com/inward/record.url?scp=85006006315&partnerID=8YFLogxK
U2 - 10.1016/j.cpc.2016.10.005
DO - 10.1016/j.cpc.2016.10.005
M3 - Article
AN - SCOPUS:85006006315
VL - 212
SP - 47
EP - 54
JO - Computer Physics Communications
JF - Computer Physics Communications
SN - 0010-4655
ER -
ID: 77960151