Power Models Database :zap:

January 17, 2024 · View on GitHub

License: ODbL v1.0

In each folder, you'll find a JSON file describing the power model of the device or component. Often a linear and a polynomial regression model.

Computing Devices

Raspberry Pi

ModelCodenameComponentModel Variable
Model Zero W (rev 1.1), 32-bit OSrbpzw1.1CPUCPU usage
Model 1 B (rev 2), 32-bit OSrbp1b2CPUCPU usage
Model 1 B+ (rev 1.2), 32-bit OSrbp1b+1.2CPUCPU usage
Model 2 B (rev 1.1), 32-bit OSrbp2b1.1CPUCPU usage
Model 3 B (rev 1.2), 32-bit OSrbp3b1.2CPUCPU usage
Model 3 B+ (rev 1.3), 32-bit OSrbp3b+1.3CPUCPU usage
Model 4 B (rev 1.1), 32-bit OSrbp4b1.1CPUCPU usage
Model 4 B (rev 1.1), 64-bit OSrbp4b1.1-64CPUCPU usage
Model 4 B (rev 1.2), 32-bit OSrbp4b1.2CPUCPU usage
Model 4 B (rev 1.2), 64-bit OSrbp4b1.2-64CPUCPU usage
Model 400 (rev 1.0), 64-bit OSrbp4001.0-64CPUCPU usage
Model 5 B (rev 1.0), 64-bit OSrbp5b1.0-64CPUCPU usage

Asus Tinker Board

ModelCodenameComponentModel Variable
Asus Tinker Board (S)asustbsCPUCPU usage

Other Hardware

Monitors

ModelCodenameComponentModel Variable
Dell P2722Hp2722hEntire deviceBrightness
HP P22Hp22hEntire deviceBrightness

Use Cases

These models are used and implemented in multiple projects and software, including:

:newspaper: License

The power models and database are licensed under the Open Data Commons Open Database License (ODbL) v1.0 (ODbL-1.0).

Copyright (c) 2022-2023, Université de Pau et des Pays de l'Adour.

Initial models and approach of Raspberry Pi devices by Adel Noureddine and Houssam Kanso, and published in the journal article: Automated Power Modeling of Computing Devices: Implementation and Use Case for Raspberry Pis.