https://github.com/rhpvorderman/sacct-analyze
Simple python script to quickly parse sacct results into max rss etc.
https://github.com/rhpvorderman/sacct-analyze
sacct slurm
Last synced: 2 months ago
JSON representation
Simple python script to quickly parse sacct results into max rss etc.
- Host: GitHub
- URL: https://github.com/rhpvorderman/sacct-analyze
- Owner: rhpvorderman
- License: mit
- Created: 2025-01-13T08:09:30.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2025-01-13T13:31:00.000Z (5 months ago)
- Last Synced: 2025-01-25T23:40:54.334Z (4 months ago)
- Topics: sacct, slurm
- Language: Python
- Homepage:
- Size: 14.6 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# sacct-analyze
The script calls `sacct --json` and parses it.
It then outputs a tab-separated table with the requested cpu, memory and time and their actual usage.
This helps to tune your request size and get better scheduling.
For example, when available cputime is much higher than actual cpu time, your job has not used all provided
cores at 100% and could probably do with less.
## Usage
```
usage: sacct-analyze.py [-h] [--name NAME]
optional arguments:
-h, --help show this help message and exit
--name NAME Should have these contents in the name
```
### Example
```
python3 sacct-analyze.py > results.tsv
```
## Example output
Name | Requested CPU | Available CPU time (hours) | Used CPU time (hours)|CPU Efficiency (%)|Requested Memory (GiB)|Used Memory (GiB)|Memory efficiency (%)|Requested Time (hours)|Used Time (hours)|Time Efficiency (%)
--- | --: | --: | --: | --: | --: | --: | --: | --: | --: | --:
call-deepVariantTask-2|4|17.83|17.13|96.03%|48.00|8.03|16.73%|83.33|4.46|5.35%
call-deepVariantTask-6|4|17.30|16.45|95.09%|48.00|7.98|16.63%|83.33|4.33|5.19%
call-deepVariantTask-1|4|50.05|48.72|97.33%|48.00|9.30|19.38%|83.33|12.51|15.02%
call-deepVariantTask-3|4|50.53|49.26|97.49%|48.00|9.39|19.57%|83.33|12.63|15.16%
call-deepVariantTask-0|4|40.17|39.08|97.29%|48.00|8.04|16.74%|83.33|10.04|12.05%
call-deepVariantTask-4|4|35.69|34.60|96.94%|48.00|8.06|16.79%|83.33|8.92|10.71%
In this example all the four provided cores are kept busy during the job runtime.
The amount of memory and time needed was much less than initially anticipated however.