woo-batch - Online in the Cloud

PROGRAM:

NAME

usage: - manual page for usage: woo-batch [-h] [-j NUM] [--job-threads NUM] [--force-
threads]

DESCRIPTION

usage: woo-batch [-h] [-j NUM] [--job-threads NUM] [--force-threads]

[--log FORMAT] [--global-log FILE] [-l LIST] [--results RESULTSDB] [--nice NICE]
[--cpu-affinity] [--executable FILE] [--rebuild] [--debug] [--gnuplot FILE]
[--dry-run] [--http-wait] [--exit-prompt] [--plot-update TIME] [--plot-timeout
TIME] [--refresh TIME] [--timing COUNT] [--timing-output FILE] [--randomize]
[--no-table] ...

Woo: batch system: runs Woo simulation multiple times with different parameters. See
https://yade-dem.org/sphinx/user.html#batch-queuing-andexecution-woo-batch for details.
Batch can be specified either with parameter table TABLE (must not end in .py), which is
either followed by exactly one SIMULATION.py (must end in .py), or contains !SCRIPT column
specifying the simulation to be run. The second option is to specify multiple scripts,
which can optionally have /nCores suffix to specify number of cores for that particular
simulation (corresponds to !THREADS column in the parameter table), e.g. sim.py/3.

positional arguments:
simulations

optional arguments:
-h, --help
show this help message and exit

-j NUM, --jobs NUM
Maximum number of simultaneous threads to run (default: number of cores, further
limited by OMP_NUM_THREADS if set by the environment: 6)

--job-threads NUM
Default number of threads for one job; can be overridden by per-job with !THREADS
(or !OMP_NUM_THREADS) column. Defaults to 1.

--force-threads
Force jobs to not use more cores than the maximum (see \-j), even if !THREADS
colums specifies more.

--log FORMAT
Format of job log files: must contain a $, % or @, which will be replaced by script
name, line number or by title column respectively. Directory for logs will be
created automatically. (default: logs/[email protected])

--global-log FILE
Filename where to redirect output of woo-batch itself (as opposed to \-\-log); if
not specified (default), stdout/stderr are used

-l LIST, --lines LIST
Lines of TABLE to use, in the format 2,3-5,8,11-13 (default: all available lines in
TABLE)

--results RESULTSDB
File (HDF5 or SQLite) where simulation should store its results (such as
input/output files and some data); the default is to use {tableName}.hdf5
({tableName}.sqlite under Windows), if there is a param table, otherwise each
simulation defines its own default files to write results in. The preferred format
is HDF5 (usually *.hdf5, *.h5, *.he5, *.hdf), SQLite is used for *.sqlite, *.db.

--nice NICE
Nice value of spawned jobs (default: 10)

--cpu-affinity
Bind each job to specific CPU cores; cores are assigned in a quasi-random order,
depending on availability at the moment the jobs is started. Each job can override
this setting by setting AFFINE column.

--executable FILE
Name of the program to run (default: /usr/bin/woo). Jobs can override with !EXEC
column.

--rebuild
Run executable(s) with --rebuild prior to running any jobs.

--debug
Run the executable with --debug. Can be overriddenn per-job with !DEBUG column.

--gnuplot FILE
Gnuplot file where gnuplot from all jobs should be put together

--dry-run
Do not actually run (useful for getting gnuplot only, for instance)

--http-wait
Do not quit if still serving overview over http repeatedly

--exit-prompt
Do not quit until a key is pressed in the terminal (useful for reviewing plots
after all simulations finish).

--plot-update TIME
Interval (in seconds) at which job plots will be updated even if not requested via
HTTP. Non-positive values will make the plots not being updated and saved unless
requested via HTTP (see \-\-plot-timeout for controlling maximum age of those).
Plots are saved at exit under the same name as the log file, with the .log
extension removed. (default: 120 seconds)

--plot-timeout TIME
Maximum age (in seconds) of plots served over HTTP; they will be updated if they
are older. (default: 30 seconds)

--refresh TIME
Refresh rate of automatically reloaded web pages (summary, logs, ...).

--timing COUNT
Repeat each job COUNT times, and output a simple table with
average/variance/minimum/maximum job duration; used for measuring how various
parameters affect execution time. Jobs can override the global value with the
!COUNT column.

--timing-output FILE
With --timing, save measured durations to FILE, instead of writing to standard
output.

--randomize
Randomize job order (within constraints given by assigned cores).

--no-table
Treat all command-line argument as simulations to be run, either python scripts or
saved simulations.

[--log FORMAT] [--global-log FILE] [-l LIST] [--results RESULTSDB] [--nice NICE]
[--cpu-affinity] [--executable FILE] [--rebuild] [--debug] [--gnuplot FILE]
[--dry-run] [--http-wait] [--exit-prompt] [--plot-update TIME] [--plot-timeout
TIME] [--refresh TIME] [--timing COUNT] [--timing-output FILE] [--randomize]
[--no-table] ...

woo-batch: error: unrecognized arguments: --version

Use woo-batch online using onworks.net services