A Sensor Predictive Model for Power Consumption using Machine Learning
Bibliographic record
Abstract
Reducing the power consumption of computing devices remains a challenge for the data center industry. In 2022, it represents approximately 2% of global electricity consumption and 1% of global greenhouse gas emissions. In addition, data centers must integrate the 5G and B5G challenges into their strategies, by increasing the computing resources available to face higher-quality service constraints. Indeed, 5G and B5G future networks are increasingly software-oriented and therefore, rely heavily on cloud computing to process large amounts of data from multiple sources in real-time.Several research works on energy management have been proposed to ensure a reduction of the energy consumed by the various components of a data center (e.g., software, computing devices, or cooling systems). However, to optimize the energy consumption of computing devices (e.g., virtual machines/container operations), it is essential to have an accurate model for predicting power consumption. Thus, we propose in this study a new sensor predictive model to predict the dynamic power consumption of cloud computing devices with high accuracy.Our proposal takes advantage of the various sensors that are now embedded in physical machines, or more generally in cloud server machines, as well as Performance Monitoring Counters to implement a Machine Learning power prediction model.The performance evaluation results confirm that our power consumption prediction models outperform previous literature models in terms of accuracy. Indeed, our best model achieves a R2score of 93.6% which is higher than the compared baseline model by 21.1%.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".