PSU Energy - AC - XGBoost - Machine

What it does #

It estimates the total machine energy consumption (AC Power) based on training data from the SPECPower database. The underlying XGBoost model can be found on our GitHub.

Classname #

• PsuEnergyAcXgboostMachineProvider

Metric Name #

• psu_energy_ac_xgboost_machine

Prerequisites & Installation #

This provider is included as a submodule in the Green Metrics Tool and should have been checked out with the initial install command of this manual. If not, run:

git submodule update --init

If you don’t have them already, you need to install some python libraries:

python3 -m pip install -r ~/green-metrics-tool/metric_providers/psu/energy/ac/xgboost/machine/model/requirements.txt

The provider must be configured in the config.yml. It must be supplied with the machine params in the config.yml file:

• CPUChips
• HW_CPUFreq
• CPUCores
• TDP
• HW_MemAmountGB

You can find these parameters in your data sheet of the used machine. In case you are using a VM please check this repository for an example how to derive the values.

Please see Configuration → for further info.

In the config.yml file the CpuUtilizationProcfsSystemProvider must also be activated for the PsuEnergyAcXgboostMachineProvider to work.

Input Parameters #

• args
  • Takes no arguments

If you want to run the provider directly we advise that you rather check out it’s main repository: XGBoost SPECPower Model documentation

The provider reads the /tmp/green-metrics-tool/cpu_utilization_procfs_system.log file from the CpuUtilizationProcfsSystemProvider in order to keep overhead low and not to double query the utilization from the system.

Output #

Since this provider should not be run directly there it has no direct output.

The resulting data however is the wattage for the whole machine (AC Power) in Watts.

This value has the same granularity as the one configured in the config.yml for the CpuUtilizationProcfsSystemProvider

Any errors are printed to Stderr.