Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GreenBytes: Intelligent Energy Estimation for Edge-Cloud

Kasra Kassai, Tasos Dagiuklas, Satwat Bashir, Muddesar Iqbal

TL;DR

GreenBytes addresses energy estimation in edge-cloud Kubernetes clusters by comparing LSTM and Gradient Booster models. The approach preprocesses CPU metrics from monitoring tools and evaluates both models on master and worker nodes to capture temporal and non-linear energy patterns. Results show the LSTM achieves lower $MSE$ and closely tracks actual energy usage, while Gradient Booster offers robust predictions across varying workloads, indicating complementary strengths. Integrating these predictors into energy management systems can enable dynamic, workload-aware power budgeting and advance sustainable computing practices in distributed environments.

Abstract

This study investigates the application of advanced machine learning models, specifically Long Short-Term Memory (LSTM) networks and Gradient Booster models, for accurate energy consumption estimation within a Kubernetes cluster environment. It aims to enhance sustainable computing practices by providing precise predictions of energy usage across various computing nodes. Through meticulous analysis of model performance on both master and worker nodes, the research reveals the strengths and potential applications of these models in promoting energy efficiency. The LSTM model demonstrates remarkable predictive accuracy, particularly in capturing dynamic computing workloads over time, evidenced by low mean squared error (MSE) rates and the ability to closely track actual energy consumption trends. Conversely, the Gradient Booster model showcases robustness and adaptability across different computational environments, despite slightly higher MSE values. The study underscores the complementary nature of these models in advancing sustainable computing practices, suggesting their integration into energy management systems could significantly enhance environmental sustainability in technology operations.

GreenBytes: Intelligent Energy Estimation for Edge-Cloud

TL;DR

GreenBytes addresses energy estimation in edge-cloud Kubernetes clusters by comparing LSTM and Gradient Booster models. The approach preprocesses CPU metrics from monitoring tools and evaluates both models on master and worker nodes to capture temporal and non-linear energy patterns. Results show the LSTM achieves lower and closely tracks actual energy usage, while Gradient Booster offers robust predictions across varying workloads, indicating complementary strengths. Integrating these predictors into energy management systems can enable dynamic, workload-aware power budgeting and advance sustainable computing practices in distributed environments.

Abstract

This study investigates the application of advanced machine learning models, specifically Long Short-Term Memory (LSTM) networks and Gradient Booster models, for accurate energy consumption estimation within a Kubernetes cluster environment. It aims to enhance sustainable computing practices by providing precise predictions of energy usage across various computing nodes. Through meticulous analysis of model performance on both master and worker nodes, the research reveals the strengths and potential applications of these models in promoting energy efficiency. The LSTM model demonstrates remarkable predictive accuracy, particularly in capturing dynamic computing workloads over time, evidenced by low mean squared error (MSE) rates and the ability to closely track actual energy consumption trends. Conversely, the Gradient Booster model showcases robustness and adaptability across different computational environments, despite slightly higher MSE values. The study underscores the complementary nature of these models in advancing sustainable computing practices, suggesting their integration into energy management systems could significantly enhance environmental sustainability in technology operations.
Paper Structure (19 sections, 6 figures)

This paper contains 19 sections, 6 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Data Preprocessing
  5. LSTM Model
  6. Gradient Booster Model
  7. Experiment Setup and data pipeline
  8. Test Bed Setup
  9. Data Collection and Storage
  10. Results and Analysis:
  11. LSTM Model Performance
  12. Master Node
  13. Worker Nodes
  14. Gradient Booster Model Performance
  15. Master Node
  16. ...and 4 more sections

Figures (6)

  • Figure 1: Actual vs Predicted Consumption on Master node running LSTM model
  • Figure 2: Actual vs Prediction Consumption for worker 1 using LSTM model trained on Master node
  • Figure 3: Actual vs Prediction Consumption for worker 2 using LSTM model trained on Master node
  • Figure 4: Actual vs Predicted Consumption on Master node running Gradient Booster model
  • Figure 5: Actual vs Prediction Consumption for worker 1 using Gradient Booster model trained on Master node
  • ...and 1 more figures