| ... | ... | @@ -491,3 +491,65 @@ optimization policies based on monitored workload. |
|
|
|
|
|
|
|
We highly recommend you to read [EARL](EARL) documentation and also how energy policies work
|
|
|
|
in order to better understand what is doing the Library internally, so you will can explore easily all features (e.g., tunning variables, collecting data) EAR offers to the end-user so you will have more knowledge about how much resources your application consumes and how to correlate with its computational characteristics.
|
|
|
|
|
|
|
|
# Data visualization with Grafana
|
|
|
|
|
|
|
|
EAR data can be visualized with Grafana dashboards in two different ways: Using grafana with SQL queries (depending on your Data Center configuration) and visualizing data collected with `eacct` and loading locally.
|
|
|
|
The second option will be explained since you might expect to not having access to the EAR Database.
|
|
|
|
|
|
|
|
Once you have your own Grafana instance running, you need to install [*csv-datasource*](https://grafana.com/grafana/plugins/marcusolsson-csv-datasource/):
|
|
|
|
|
|
|
|
```bash
|
|
|
|
bin/grafana-cli plugins install marcusolsson-csv-datasource (You can first check if it's already available by testing the available Data sources)
|
|
|
|
```
|
|
|
|
|
|
|
|
Enable the CSV plug-in by creating a `custom.ini` file in the conf directory with the following content:
|
|
|
|
|
|
|
|
```ini
|
|
|
|
[plugin.marcusolsson-csv-datasource]
|
|
|
|
allow_local_mode = true
|
|
|
|
```
|
|
|
|
|
|
|
|
Once you have a local server running on your PC or laptop, open your web browser and connect to Grafana at the URL: `http://localhost:3000/login`.
|
|
|
|
Next steps are:
|
|
|
|
|
|
|
|
**Create the Data source**
|
|
|
|
|
|
|
|
In the left menu, select Configuration/Data source/Add data source. Select CSV data source from the list of options.
|
|
|
|
You need to create a new data source for each CSV file you are going to visualize.
|
|
|
|
For each one, select *Local*. Note that the path must to be a **public directory**.
|
|
|
|
|
|
|
|
**Import the Dashboard**
|
|
|
|
|
|
|
|
Go to the left menu, *Dashboard*, and select the *Import* option.
|
|
|
|
This option allows you uploading or selecting a json file with pre-specified graphs, tables, etc.
|
|
|
|
Graphs are associated with data sources, so you may need to change the Data Source name in the json file to match the one you've created on Grafana.
|
|
|
|
The json file is [here](misc/EAR_job_data_visualization.json), and below you can see the Data Source names expected.
|
|
|
|
There is a configuration for two data sources: *EAR_loops* for visualizing CSV files containing EAR loop signatures (e.g., `eacct [-j <job_id>[.<step_id>]] -r -c <filename>`) and *EAR_app* for visualizing application signatures (e.g., `eacct [-j <job_id>[.<step_id>]] -l -c <filename>`).
|
|
|
|
|
|
|
|
```json
|
|
|
|
{
|
|
|
|
"__inputs": [
|
|
|
|
{
|
|
|
|
"name": "DS_EAR_LOOPS",
|
|
|
|
"label": "EAR_loops",
|
|
|
|
"description": "",
|
|
|
|
"type": "datasource",
|
|
|
|
"pluginId": "marcusolsson-csv-datasource",
|
|
|
|
"pluginName": "CSV"
|
|
|
|
},
|
|
|
|
{
|
|
|
|
"name": "DS_EAR_APPS",
|
|
|
|
"label": "EAR_apps",
|
|
|
|
"description": "",
|
|
|
|
"type": "datasource",
|
|
|
|
"pluginId": "marcusolsson-csv-datasource",
|
|
|
|
"pluginName": "CSV"
|
|
|
|
}
|
|
|
|
],
|
|
|
|
```
|
|
|
|
|
|
|
|
Import the JSON file to create the visualization dashboards and refresh the URL the browser page.
|
|
|
|
Below you can see an example of what you will see.
|
|
|
|
|
|
|
|
 |
