Sawmill is a SchedMD SLURM log analyzer (it also supports the 1021 other log formats listed to the left).
It can process log files
in SchedMD SLURM format, and generate dynamic statistics from them,
analyzing and reporting events.
Sawmill can parse SchedMD SLURM logs, import them into a MySQL, Microsoft SQL Server, or Oracle database (or its own built-in database),
aggregate them, and generate dynamically filtered reports, all through a web interface.
Sawmill can perform SchedMD SLURM log analysis on any platform, including Windows, Linux, FreeBSD, OpenBSD, Mac OS, Solaris, other UNIX, and others.
Sawmill stores the following non-numerical fields in its database for SchedMD SLURM, generates reports for each field, and allows dynamic filtering on any combination of these fields:
|Field|| ||Internal Name|
| ||date/time|| ||date_time|
| ||day of week|| ||day_of_week|
| ||hour of day|| ||hour_of_day|
| ||account|| ||account|
| ||eligible time|| ||eligible|
| ||end time|| ||end|
| ||job ID|| ||jobid|
| ||job name|| ||jobname|
| ||partition|| ||partition|
| ||quality of service|| ||qos|
| ||start time|| ||start|
| ||submission time|| ||submit|
| ||time limit|| ||timelimit|
| ||user|| ||user|
Sawmill stores the following numerical fields in its database for SchedMD SLURM, aggregating them and including them as columns in most reports:
|Numerical Field|| ||Internal Name|
| ||events|| ||events|
| ||CPU time|| ||cputimeraw|
| ||CPUs|| ||ncpus|
| ||nodes|| ||nnodes|
| ||wait time|| ||wait_time|
See Sawmill Features to learn more about Sawmill's options for viewing, customizing, filtering, exporting and scheduling SchedMD SLURM reports.
Sawmill also supports 1021 other log formats.