1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
|
Input parameters
================
.. warning::
This section is currently in development.
Overall simulation parameters
-----------------------------
* ``max_step`` (`integer`)
The number of PIC cycles to perform.
Setting up the field mesh
-------------------------
* ``amr.n_cell`` (`2 integers in 2D`, `3 integers in 3D`)
The number of grid points along each direction (on the **coarsest level**)
* ``amr.max_level`` (`integer`)
When using mesh refinement, the number of refinement levels that will be used.
Use 0 in order to disable mesh refinement.
* ``geometry.is_periodic`` (`2 integers in 2D`, `3 integers in 3D`)
Whether the boundary conditions are periodic, in each direction.
For each direction, use 1 for periodic conditions, 0 otherwise.
* ``geometry.prob_lo`` and ``geometry.prob_hi`` (`2 floats in 2D`, `3 integers in 3D`; in meters)
The extent of the full simulation box. This box is rectangular, and thus its
extent is given here by the coordinates of the lower corner (``geometry.prob_lo``) and
upper corner (``geometry.prob_hi``).
* ``warpx.fine_tag_lo`` and ``warpx.fine_tag_hi`` (`2 floats in 2D`, `3 integers in 3D`; in meters)
**When using static mesh refinement with 1 level**, the extent of the refined patch.
This patch is rectangular, and thus its extent is given here by the coordinates
of the lower corner (``warpx.fine_tag_lo``) and upper corner (``warpx.fine_tag_hi``).
Distribution across MPI ranks and parallelization
-------------------------------------------------
* ``amr.max_grid_size`` (`integer`)
Maximum allowable size of each **subdomain**
(expressed in number of grid points, in each direction).
Each subdomain has its own ghost cells, and can be handled by a
different MPI rank.
If ``max_grid_size`` is such that the total number of subdomains is
**larger** that the number of MPI ranks used, than some MPI ranks
will handle several subdomains, thereby providing additional flexibility
for **load balancing**.
When using mesh refinement, this number applies to the subdomains
of the coarsest level, but also to any of the finer level.
* ``warpx.load_balance_int`` (`integer`)
How often WarpX should try to redistribution the work across MPI ranks,
in order to have better load balancing (expressed in number of PIC cycles
inbetween two consecutive attempts at redistributing the work).
Use 0 to disable load_balancing.
When performing load balancing, WarpX measures the wall time for
computational parts of the PIC cycle. It then uses this data to decide
how to redistribute the subdomains across MPI ranks. (Each subdomain
is unchanged, but its owner is changed in order to have better performance.)
This relies on each MPI rank handling several (in fact many) subdomains
(see ``max_grid_size``).
Particle initialization
-----------------------
Laser initialization
--------------------
* ``warpx.use_laser`` (`0 or 1`)
Whether to activate the injection of a laser pulse in the simulation
* ``laser.profile`` (`string`)
The spatio-temporal shape of the laser. The options that are currently
implemented are:
- ``"Gaussian"``: The transverse and longitudinal profiles are Gaussian.
- ``"Harris"``: The transverse profile is Gaussian, but the longitudinal profile is given by the Harris function (see ``laser.profile_duration`` for more details)
* ``laser.e_max`` (`float` ; in V/m)
Peak amplitude of the laser field.
For a laser with a wavelength :math:`\lambda = 0.8\,\mu m`, the peak amplitude
is related to :math:`a_0` by:
.. math::
E_{max} = a_0 \frac{2 \pi m_e c}{e\lambda} = a_0 \times (4.0 \cdot 10^{12} \;V.m^{-1})
* ``laser.position`` (`3 floats in 3D and 2D` ; in meters)
The coordinates of one of the point of the antenna that will emit the laser.
The plane of the antenna is entirely defined by ``laser.position`` and ``laser.direction``.
``laser.position`` also corresponds to the origin of the coordinates system
for the laser tranverse profile. For instance, for a Gaussian laser profile,
the peak of intensity will be at the position given by ``laser.position``.
This variable can thus be used to shift the position of the laser pulse
transversally.
.. note::
In 2D, ``laser.position`` is still given by 3 numbers, but the second number is ignored.
* ``laser.profile_t_peak`` (`float`; in seconds)
The time at which the laser reaches its peak intensity, at the position
given by ``laser.position`` (only used for the ``"gaussian"`` profile)
* ``laser.profile_duration`` (`float` ; in seconds)
The duration of the laser, defined as :math:`\tau` below:
- For the ``"gaussian"`` profile:
.. math::
E(\boldsymbol{x},t) \propto \exp\left( -\frac{(t-t_{peak})^2}{\tau^2} \right)
- For the ``"harris"`` profile:
.. math::
E(\boldsymbol{x},t) \propto \frac{1}{32}\left[10 - 15 \cos\left(\frac{2\pi t}{\tau}\right) + 6 \cos\left(\frac{4\pi t}{\tau}\right) - \cos\left(\frac{6\pi t}{\tau}\right) \right]
* ``laser.profile_waist`` (`float` ; in meters)
The waist of the transverse Gaussian laser profile, defined as :math:`w_0` :
.. math::
E(\boldsymbol{x},t) \propto \exp\left( -\frac{\boldsymbol{x}_\perp^2}{w_0^2} \right)
For :math:`t>\tau`, :math:`E(\boldsymbol{x},t)=0`.
* ``laser.wavelength`` (`float`; in meters)
The wavelength of the laser in vacuum.
* ``laser.polarization`` (`3 floats in 3D and 2D`)
The coordinates of a vector that points in the direction of polarization of
the laser. The norm of this vector is unimportant, only its direction matters.
.. note::
Even in 2D, all the 3 components of this vectors are important (i.e.
the polarization can be orthogonal to the plane of the simulation).
* ``laser.direction`` (`3 floats in 3D`)
The coordinates of a vector that points in the propagation direction of
the laser. The norm of this vector is unimportant, only its direction matters.
The plane of the antenna that will emit the laser is orthogonal to this vector.
* ``laser.profile_focal_distance`` (`float`; in meters)
The distance from ``laser_position`` to the focal plane.
(where the distance is defined along the direction given by ``laser.direction``.)
Use a negative number for a defocussing laser instead of a focussing laser.
Numerics and algorithms
-----------------------
* ``warpx.use_filter`` (`0 or 1`)
Whether to smooth the charge and currents on the mesh, after depositing
them from the macroparticles. This uses a bilinear filter
(see the sub-section **Filtering** in :doc:`../theory/theory`).
* ``algo.current_deposition`` (`integer`)
The algorithm to be used for current deposition:
- ``0``: Esirkepov deposition, vectorized
- ``1``: Esirkepov deposition, non-optimized
- ``2``: Direct deposition, vectorized
- ``3``: Direct deposition, non-optimized
* ``algo.charge_deposition`` (`integer`)
The algorithm to be used for the charge density deposition:
- ``0``: Vectorized version
- ``1``: Non-optimized version
* ``algo.field_gathering`` (`integer`)
The algorithm to be used for field gathering:
- ``0``: Vectorized version
- ``1``: Non-optimized version
* ``algo.particle_pusher`` (`integer`)
The algorithm to be used for the particle pusher:
- ``0``: Boris pusher
- ``1``: Vay pusher
Diagnostics and output
----------------------
* ``amr.plot_int`` (`integer`)
The number of PIC cycles inbetween two consecutive data dumps. Use a
negative number to disable data dumping.
|
