Usage guide¶
Synopsis¶
hpcstats [-h –help] [-c –conf pathname] [-d –debug] action options
Description¶
HPCStats is a software to manage a relational database which centralizes the production usage data of various HPC supercomputers. This command can import new production data, check data sources availability, modify the content of the database and generate usage reports with statistics.
Actions¶
Available actions:
- import
- Import new production data and update database incrementally.
- check
- Check production data sources availability.
- modify
- Modify the content of the database.
- report
- Generate usage report with statistics.
Global options¶
-h, --help Show help message and exit. -c FILE, --conf=FILE Alternative configuration file. Default is /etc/hpcstats/hpcstats.conf. -d, --debug Enable debug output. --batch-mode Set output in batch mode. The outputs are sent to local syslog server, the standard and errors outputs only get the messages at the warning level and over. This mode is particularly usefull for cronjobs because the warnings and errors are sent by email when needed and administrators get all output in syslog for further diagnosis.
Import options¶
Required arguments for import action:
--cluster=CLUSTER The name of the cluster from which the data will be imported. Special value all can be used to import data from all configured supercomputers. —since-event=DATE Import events starting from this date for new clusters. The date must be formatted like YYYY-MM-DD. Default is 1970-01-01.
--since-jobid=ID Import jobs starting from this internal job id for new clusters. The job id must an integer. Default is -1.
Check options¶
Required arguments for check action:
--cluster=CLUSTER The name of the cluster for which the data sources will be checked. Special value all can be used to check data sources for all configured supercomputers.
Modify options¶
Optional arguments for modify action:
--business-code=CODE The business code to modify. --project-code=CODE The project code to modify. --set-description=DESC The new description of the business code or the project. --set-domain=DOMAIN The new domain of the project. --new-domain=DOMAIN The new domain key to create in database. --domain-name=NAME The new domain name to create in database.
Please refer to Examples section to see the possible combinations of the parameters.
Report options¶
Disclaimer: This feature must be considered as very experimental!
Required arguments for report action:
--cluster=CLUSTER The name of the cluster for which the report is generated.
Optional arguments for report action:
--interval=ITL The interval used in the report. Default is day. --template=TPL The template to use for the report. Default is csv.
Examples¶
Import production data of supercomputer HPC-A:
hpcstats import --cluster=HPC-A
Import production data of all supercomputers:
hpcstats import --cluster=all
Check supercomputer HPC-B data sources availablity:
hpcstats check --cluster=HPC-B
Set new description to business code B1:
hpcstats modify --business-code=B1 --set-description='new description B1'
Set new description to project P1:
hpcstats modify --project-code=P1 --set-description='new description P1'
Set new domain D1 to project P1:
hpcstats modify --project-code=P1 --set-domain=D1
Set create new domain D2:
hpcstats modify --new-domain=D2 --domain-name='domain name D2'