Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/dcoles/prometheus-pandas
Pandas integration for Prometheus.
https://github.com/dcoles/prometheus-pandas
data-analysis jupyter-notebook pandas prometheus python
Last synced: 11 days ago
JSON representation
Pandas integration for Prometheus.
- Host: GitHub
- URL: https://github.com/dcoles/prometheus-pandas
- Owner: dcoles
- License: mit
- Created: 2018-11-26T04:32:17.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2024-09-13T17:32:56.000Z (about 2 months ago)
- Last Synced: 2024-10-13T13:32:38.077Z (24 days ago)
- Topics: data-analysis, jupyter-notebook, pandas, prometheus, python
- Language: Jupyter Notebook
- Homepage: https://pypi.org/project/prometheus-pandas/
- Size: 190 KB
- Stars: 46
- Watchers: 3
- Forks: 9
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Prometheus Pandas
Python library for querying [Prometheus](https://prometheus.io/) and accessing the results as
[Pandas](https://pandas.pydata.org/) data structures.
This is mostly intended for use in [Jupyter](https://jupyter.org/) notebooks. See [Prometheus.ipynb](Prometheus.ipynb) for an example.
## Example
Evaluate an instant query at a single point in time:
```python
>>> from prometheus_pandas import query
>>>
>>> p = query.Prometheus('http://localhost:9090')
>>> series = p.query('node_cpu_seconds_total{mode="system"}', '2020-05-10T00:00:00Z')
>>> print(series)
node_cpu_seconds_total{cpu="0",instance="localhost:9100",job="node",mode="system"} 15706.47
node_cpu_seconds_total{cpu="1",instance="localhost:9100",job="node",mode="system"} 15133.25
node_cpu_seconds_total{cpu="2",instance="localhost:9100",job="node",mode="system"} 15095.59
node_cpu_seconds_total{cpu="3",instance="localhost:9100",job="node",mode="system"} 14649.20
dtype: float64
```
Evaluates an expression query over a time range:
```python
>>> from prometheus_pandas import query
>>>
>>> p = query.Prometheus('http://localhost:9090')
>>> dataframe = p.query_range(
'sum(rate(node_cpu_seconds_total{mode=~"system|user"}[1m])) by (mode)',
'2020-10-05T00:00:00Z', '2020-10-05T06:00:00Z', '1h')
>>> print(dataframe)
dtype: float64
---
{mode="system"} {mode="user"}
2020-10-05 00:00:00 0.022667 0.038222
2020-10-05 01:00:00 0.015333 0.036667
2020-10-05 02:00:00 0.028000 0.040667
2020-10-05 03:00:00 0.015111 0.034889
2020-10-05 04:00:00 0.015556 0.038000
2020-10-05 05:00:00 0.018444 0.040222
2020-10-05 06:00:00 0.018222 0.035111
```
Customizing the HTTP request:
```python
>>> import requests
>>> from prometheus_pandas import query
>>>
>>> http = requests.Session()
>>> http.auth = ('user', 'pass') # Basic authentication
>>> http.cert = '/path/client.cert' # X.509 client certificate authentication
>>> http.verify = '/path/to/certfile' # Custom certificate bundle
>>>
>>> p = query.Prometheus('http://localhost:9090', http)
>>> series = p.query('node_cpu_seconds_total{mode="system"}', '2020-10-05T00:00:00Z')
>>> print(series)
node_cpu_seconds_total{cpu="0",instance="localhost:9100",job="node",mode="system"} 3954.92
dtype: float64
```
## Installation
Latest release via [`pip`](https://pip.pypa.io):
```bash
pip install prometheus-pandas [--user]
```
via Git:
```bash
git clone https://github.com/dcoles/prometheus-pandas.git; cd prometheus-pandas
python3 setup.py install [--user]
```
## Licence
Licenced under the [MIT License](https://choosealicense.com/licenses/mit/). See `LICENSE` for details.