Particle-particle Particle-tree Code for Planetary System Formation with Individual Cut-off Method: GPLUM. (arXiv:2007.15432v1 [astro-ph.EP])
<a href="http://arxiv.org/find/astro-ph/1/au:+Ishigaki_Y/0/1/0/all/0/1">Yota Ishigaki</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Kominami_J/0/1/0/all/0/1">Junko Kominami</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Makino_J/0/1/0/all/0/1">Junichiro Makino</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Fujimoto_M/0/1/0/all/0/1">Masaki Fujimoto</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Iwasawa_M/0/1/0/all/0/1">Masaki Iwasawa</a>
In the standard theory of the formation of planets in our Solar System,
terrestrial planets and the cores of gas giant planets are formed through
accretion of km size objects (planetesimals) in the protoplanetary disk. The
gravitational $N$-body simulations of planetesimal systems is the most direct
way to study the accretion process of the planets. However, the use of $N$-body
simulations has been limited to idealized model (e.g., perfect accretion)
and/or narrow radial range, due to the limited number of particles available.
We have developed a new N-body simulation code with particle-particle
particle-tree (P3T) scheme for planetary system formation, GPLUM. GPLUM uses a
fourth-order Hermite scheme to calculate gravitational interactions between
particles within cut-off radii of individual particles and the Barnes-Hut tree
scheme for gravitational interactions with particles outside the cut-off radii.
In existing implementations of the P3T schemes, the same cut-off radius is used
for all particles. Thus, when the range of the mass of the planetesimals
becomes large, the calculation speed decreases. We have solved this problem by
allowing each particle to determine its appropriate cutoff radius depending on
its mass, distance from the central star, and the local velocity dispersion,
and thus achieved a significant speedup when the range of the masses of
particles is wide. We have also improved the scalability of the code, and have
achieved good strong-scaling performance for up to 1,024 cores in the case of
N=10^6. GPLUM is freely available from https://github.com/YotaIshigaki/GPLUM
with MIT licence.
In the standard theory of the formation of planets in our Solar System,
terrestrial planets and the cores of gas giant planets are formed through
accretion of km size objects (planetesimals) in the protoplanetary disk. The
gravitational $N$-body simulations of planetesimal systems is the most direct
way to study the accretion process of the planets. However, the use of $N$-body
simulations has been limited to idealized model (e.g., perfect accretion)
and/or narrow radial range, due to the limited number of particles available.
We have developed a new N-body simulation code with particle-particle
particle-tree (P3T) scheme for planetary system formation, GPLUM. GPLUM uses a
fourth-order Hermite scheme to calculate gravitational interactions between
particles within cut-off radii of individual particles and the Barnes-Hut tree
scheme for gravitational interactions with particles outside the cut-off radii.
In existing implementations of the P3T schemes, the same cut-off radius is used
for all particles. Thus, when the range of the mass of the planetesimals
becomes large, the calculation speed decreases. We have solved this problem by
allowing each particle to determine its appropriate cutoff radius depending on
its mass, distance from the central star, and the local velocity dispersion,
and thus achieved a significant speedup when the range of the masses of
particles is wide. We have also improved the scalability of the code, and have
achieved good strong-scaling performance for up to 1,024 cores in the case of
N=10^6. GPLUM is freely available from https://github.com/YotaIshigaki/GPLUM
with MIT licence.
http://arxiv.org/icons/sfx.gif
