Kepler Power Model
In Kepler, with respective to available measurements, we provide a pod-level power with a mix of two power modeling approaches:
Modeling Approach
-
Power Ratio Modeling: This modeling computes a finer-grained power by the usage ratio over the total summation of power. This modeling is used by default when the total power is known.
-
Power Estimation Modeling: This modeling estimates a power by using usage metrics as input features of the trained model. This modeling can be used even if the power metric cannot be measured. The estimation can be done in three levels: Node total power (including fan, power supply, etc.), Node internal component powers (such as CPU, Memory), Pod power.
Note
Also see Get started with Kepler Model Server
-
Pre-trained Power Models: We provide pre-trained power models for different deployment scenarios. Current x86_64 pre-trained model are developed in Intel® Xeon® Processor E5-2667 v3. Models with other architectures are coming soon. You can find these models in Kepler Model DB. These models support both power ratio modeling and power estimation modeling for both RAPL and ACPI power sources. The
AbsPowermodels estimate both idle and dynamic power while theDynPowermodels only estimate dynamic power. The MAE (Mean Absolute Error) of these models are also published. Kepler container image has preloaded acpi/AbsPower/BPFOnly/SGDRegressorTrainer_1.json model for node energy estimate and rapl/AbsPower/BPFOnly/SGDRegressorTrainer_1.json for Container absolute power estimate.
Usage Scenario
| Scenario | Node Total Power | Node Component Powers | Pod Power |
|---|---|---|---|
| BM (x86 with power meter) | Measurement (e.g., ACPI) | Measurement (RAPL) | Power Ratio |
| BM (x86 but no power meter) | Power Estimation | Measurement | Power Ratio |
| BM (non-x86 with power meter) | Measurement | Power Estimation | Power Ratio |
| BM (non-x86 and no power meter) | Power Estimation | Power Estimation | Power Ratio |
| VM with node info and power passthrough from BM (x86 with power meter) | Measurement + VM Mapping | Measurement + VM Mapping | Power Ratio |
| VM with node info and power passthrough from BM (x86 but no power meter) | Power Estimation | Measurement + VM Mapping | Power Ratio |
| VM with node info and power passthrough from BM (non-x86 with power meter) | Measurement + VM Mapping | Power Estimation | Power Ratio |
| VM with node info | Power Estimation | Power Estimation | Power Ratio |
| Pure VM | - | - | Power Estimation |