Closed ArneTR closed 1 week ago
Eco-CI Output: | Label | 🖥 avg. CPU utilization [%] | 🔋 Total Energy [Joules] | 🔌 avg. Power [Watts] | Duration [Seconds] |
---|---|---|---|---|---|
Total Run (incl. overhead) | 25.0178 | 1570.25 | 3.93 | 400 | |
Measurement #1 | 24.9735 | 1570.25 | 3.95 | 398 |
🌳 CO2 Data: City: Boydton, Lat: 36.6534, Lon: -78.375 IP: 20.161.78.144 CO₂ from energy is: 0.617108250 g CO₂ from manufacturing (embodied carbon) is: 0.114125523 g Carbon Intensity for this location: 393 gCO₂eq/kWh SCI: 0.731234 gCO₂eq / pipeline run emitted
@ribalba All done. Quick Re-review please as I changed this also in all other C source files we have
This PR adds a Disk I/O reporter.
To make it container agnostic the data is pulled directly from the
sysfs
. This needs the IO delegator to be set as specified in the documentation.I am currently under the assumption that this controller is always active for docker in root mode, but will double check soon.
The approach here is similar to the other cgroup reading metric providers. However while crafting this the idea came up to actually split Input and Output and not supply the value as a cumulated number.
I see here two options:
Here I think @ribalba it makes sense to discuss how this was solved in the lm_sensors