Skip to content

Latest commit

 

History

History
22 lines (16 loc) · 2.42 KB

README.md

File metadata and controls

22 lines (16 loc) · 2.42 KB
Raw

🌍 Carbon Aware Computing for GenAI Developers

🌱 Welcome to the "Carbon Aware Computing for GenAI Developers" course! The course will equip you with the skills to perform model training and inference jobs using cleaner, low-carbon energy in the cloud.

Course Summary

In this course, you'll learn how to make environmentally conscious decisions while performing machine learning tasks, optimizing the use of clean electricity. Here's what you can expect to learn and experience:

  1. 🌐 Real-time Electricity Data: Query real-time electricity grid data to understand the power breakdown (e.g., wind, hydro, coal) and carbon intensity (CO2 equivalent emissions per kWh) of various regions.
  2. ⚡ Low-carbon Model Training: Train models with low-carbon energy by selecting regions with low average carbon intensity for your training jobs and data uploads. Further optimize by using real-time grid data from ElectricityMaps.
  3. 📊 Carbon Footprint Measurement: Retrieve measurements of the carbon footprint for ongoing cloud jobs using the Google Cloud Carbon Footprint tool, which estimates greenhouse gas emissions from your Google Cloud usage.
  4. 🌍 Carbon-aware Development: Throughout the course, you'll use ElectricityMaps, a free API for querying global electricity grid information, and Google Cloud to run model training jobs in data centers powered by low-carbon energy.

Key Points

  • 📡 Global Energy Data: Retrieve real-time data on global energy mixes and carbon intensity from the ElectricityMaps API, identifying power grids that produce electricity from low-carbon sources.
  • 🔄 Optimized Training Jobs: Run machine learning training jobs using low-carbon electricity by redirecting tasks to cloud server locations based on their carbon intensity measurements.
  • 📉 Carbon Footprint Analysis: Analyze the carbon footprint of sample Google Cloud usage data, including machine learning training, inference, storage, and other API activities.

About the Instructor

🌟 Nikita Namjoshi is a Developer Advocate at Google Cloud and Google Fellow on the Permafrost Discovery Gateway, bringing extensive expertise in environmentally conscious computing to guide you through this course.

🔗 To enroll in the course or for further information, visit deeplearning.ai.