Configuration¶
perun can be configured to fit different types of goals and workflows, either on the command line, decorator or using a configuration file. For more details on how use the options, go to the Usage section.
The following table list the current options available, their default values, and a brief description.
Options¶
Category |
Name |
Default |
Description |
|---|---|---|---|
post-processing |
power_overhead |
0.0 |
Estimated power consumption of non-measured hardware components in Watts. Will be added to the power draw and energy consumed of individual nodes. Defaults to 0 Watts |
post-processing |
pue |
1.0 |
Power Usage Effectiveness: A measure of a data centers efficiency, calculated as PUE = Total facilitty energy / IT equipment energy. Calculated for each run. |
post-processing |
emissions_factor |
417.80 |
Average carbon intensity of electricity (gCO2e/kWh). Calculated for each run. Source: https://ourworldindata.org/grapher/carbon-intensity-electricity |
post-processing |
price_factor |
0.3251 |
Power to Currency conversion factor (Currency/kWh). Calculated for each run. Source : https://www.stromauskunft.de/strompreise/ |
post-processing |
price_unit |
€ |
Currency Icon |
monitor |
sampling_period |
1 |
Seconds between measurements |
monitor |
include_backends |
Space separated list of backends to include during monitoring. If empty, all backends will be included. Cannot be used together with exclude_backends. |
|
monitor |
exclude_backends |
Space separated list of backends to exclude during monitoring. If empty, all backends will be included. Cannot be used together with include_backends. |
|
monitor |
include_sensors |
Space separated list of sensors to include during monitoring. If empty, all sensors will be included. Cannot be used together with exclude_sensors. |
|
monitor |
exclude_sensors |
Space separated list of sensors to exclude during monitoring. If empty, all sensors will be included. Cannot be used together with include_sensors. |
|
output |
app_name |
None |
Name to identify the app. If None, name will be based on the file or function name. |
output |
run_id |
None |
ID of the current run. If None, the current date and time will be used. If SLURM, perun will look for the environmental variable SLURM_JOB_ID and use that. |
output |
format |
text |
Output report format [text, pickle, csv, hdf5, json, bench] |
output |
data_out |
./perun_results |
perun output location. |
benchmarking |
rounds |
1 |
Number of times the application is run. |
benchmarking |
warmup_rounds |
0 |
Number of warmup rounds to run before starting the benchmarks. |
benchmarking |
metrics |
runtime,energy |
List of metrics to present on the benchmarking report. |
benchmarking |
region_metrics |
runtime,power |
List of metrics to present on the region report. |
benchmarking.units |
joule |
k |
Default order of magnitude to present total energy for the benchmarking report. |
benchmarking.units |
second |
Default order of magnitude to present runtime information for the benchmarking report. |
|
benchmarking.units |
percent |
Default order of magnitude to present percentages for the benchmarking report. |
|
benchmarking.units |
watt |
k |
Default order of magnitude to present power draw for the benchmarking report. |
benchmarking.units |
byte |
G |
Default order of magnitude to present number of bytes for the benchmarking report. |
debug |
log_lvl |
WARNING |
Change logging output [DEBUG, INFO, WARNING, ERROR, CRITICAL] |