This project aims to study how to optimize carbon emissions in cloud data centers using resource availability estimation and carbon awareness strategies. By analyzing cluster resource data, carbon intensity data, and workflow data, we explored how to reduce carbon emissions, optimize task scheduling, and improve resource utilization efficiency.
- availability_resources_estimation_1.ipynb: Jupyter Notebook comparing resource availability estimation with carbon intensity data using forecast data.
- availability_resources_estimation_2.ipynb: Jupyter Notebook comparing resource availability estimation with carbon intensity data using actual data.
- batch_task.csv: Alibaba cluster data containing batch processing jobs.
- batch_task_running.ipynb: Jupyter Notebook for analyzing cluster data and batch task running.
- carbon_intensity_202208.csv: Carbon intensity data for August 2022.
- carbon_intensity_data_processing.ipynb: Jupyter Notebook for processing carbon intensity data.
- input_trace_lotaru.csv: Workflow data for simulations.
- lotaru_data_input.ipynb: Jupyter Notebook for processing workflow data.
- online_service_free_capacity.ipynb: Alibaba cluster data for online service and free capacity analysis.
- power_consumption_estimation.ipynb: Jupyter Notebook for implementing the linear power model.
- result_emssion_saved.csv: CSV file containing simulation results on emission savings.
- simulation.ipynb: Jupyter Notebook for constructing and running the simulation system.
- README.md: Project overview, contents, and instructions.