forked from amusecode/amuse
-
Notifications
You must be signed in to change notification settings - Fork 0
/
amuserc.example
148 lines (109 loc) · 5.15 KB
/
amuserc.example
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
# This is an example configuration file for AMUSE
# AMUSE will look for the `amuserc` file in the following locations
# 1) amuse python path (src/amuse, used for binary installs)
# 2) amuse root directory (set by the AMUSE_DIR environment variable)
# 3) your home directory (under the name .amuserc, so a hidden file)
# 4) a file set in the AMUSERC environment variable
# 5) the current working directory
# AMUSE will combine the files it finds in these paths.
# The file in the path with the higher number will overwrite the values
# in the file(s) in the path with lower number. For example
# the values in the file in the current working directory will
# override the values in the file in the amuse root directory.
# Amuse will use the 'amuserc' filename, except for the home directory
# were it will look for .amuserc (${HOME}/.amuserc)
[channel]
# You can redirect the output of a code to a file, /dev/null or not at all
# by default all output is redirector to /dev/null and will not be
# shown to the running script.
# valid values are:
# none - no redirection, all output is shown on the console
# null - redirect to /dev/null, no output is shown
# file - redirect to a file
# redirection=null
# If you want to redirect to a file, you'll need to uncomment these
# lines and specify the names of the files for standard output and
# error output. The name for stderr can be the same as stdout.
# redirect_stdout_file=code.out
# redirect_stderr_file=code.out
# You can specify the code to be run under a debbugger
# valid values are
# none - no debugger, normal run
# gdb - start a gdb debugger in a separate xterm
# xterm - start the code under xterm, ideal for console logging
# ddd - start a ddd session (a graphical debugger)
# valgrind - run the code under valgrind to check for memory leaks
# gdb-remote - run the gdb as a server that waits on a port
# debugger=none
# for remote debugging you can specify a port to listen on
# debugger=gdb-remote
# debugger_port=4343
# You can specify the communication channel and process start
# library. Default value is mpi or if mpi is not available
# sockets.
# valid values are:
# mpi - use mpi library to start and run the community codes
# sockets - use simple forks and sockets, only runs on one node
# ibis - use the ibis java library to run on a hetrogeneous grid
# not all codes are compatible with sockets and ibis
# channel_type=mpi
###################################################################
# Advanced channel options
#
# Also look for executables in the specified directory,
# needs to be an absolute directory name
# (for binary installs)
# worker_code_directory=/tmp/codes/
# Maximum number of calls to merge into one call
# AMUSE is optimized to send large blocks of data between the
# script and an community code, this lowers communication
# overhead but increases memory usage. If you encounter memory
# problems (many running codes on a single node) you can
# lower this number
# max_message_length=1000000
# The redirection is implemented with a simple python script
# run with the same executable as the main python script
# for remote systems you can specify a different python executable
# python_exe_for_redirection=/usr/bin/python
# By default the community code executable modification date
# is checked against the interface.py file. If the executable
# is not up to date and error is raises. In some cases you
# may want to turn off this check.
# must_check_if_worker_is_up_to_date=True
# For MPICH2 AMUSE checks if the mpd is running for every code
# start. This prevents hangs but causes some overhead. If you
# want to start many different codes you can turn of the
# check with this option
# check_mpi=True
# On some MPI installations no new codes can be started
# after disconnecting with a previous code (code.stop()).
# It is possible to prevent disconnection during a run by
# setting this option to False.
# must_disconnect_on_stop=True
# on some MPI installations, idle workers cycle 100% cpu, if this is causing
# performance issues set the following. This increases the latency
# for communication
# polling_interval_in_milliseconds=10
###################################################################
[data]
# AMUSE will look for certain configuration files
# and data files in the amuse_root_dir. For source code installations
# the amuse code will automatically determine the correct
# directory, but for custom installations you might need to
# change it
# amuse_root_dir=/usr/share/amuse
# Some codes in AMUSE need data or parameter files to work
# correctly. By defailt AMUSE will look for `amuse_root_dir`/data
# you can change the value with this parameter
# input_data_root_directory=/usr/share/amuse/data
# Some codes in AMUSE write data while running.
# By default AMUSE will use `amuse_root_dir`/data.
# You can specify a different directory. If you give
# a relative path, a directy relative to your current working directory
# will be used. AMUSE will create the directory if it does not already
# exist.
# input_data_root_directory=.
[sockets_channel]
# when the sockets channel is used, the following may be necessary
# (experiment in case of problems)
# close_fds=False