Watt's the Best Time to Compute? Google's Road to the Edelman Award
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Watt's the Best Time to Compute? Google's Road to the Edelman Award” inside PodZeus.
This episode of Resoundingly Human highlights Google's groundbreaking carbon-aware computing system, a finalist for the 2026 Franz Edelman Award. The project, led by a cross-functional team of engineers and data scientists including Anna Radovanovic, Ross Conningstein, Valfredo Sirni, and Adrian, tackles the immense challenge of reducing the carbon footprint of Google's global data center network. By intelligently shifting flexible computing workloads across time and geography based on real-time and predictive grid carbon intensity data, the system optimizes energy use to align with cleaner power sources. The team shares the technical, organizational, and governance complexities involved in building such a large-scale system, emphasizing the importance of analytics, feedback control, and day-ahead forecasting. Over time, the system has not only reduced carbon emissions but also enabled peak load shaving, improved energy efficiency, and even supported grid resilience during emergencies like the 2021 Oklahoma storm. The team also emphasizes the project’s broader impact by open-sourcing research and tools to inspire other tech companies to adopt similar practices. The episode underscores how operations research and data science can drive sustainable innovation at scale. The team reflects on the power of collaboration, persistence, and a shared mission to create meaningful change. Their journey—from proving feasibility to gaining organizational buy-in and scaling globally—serves as an inspirational model for data scientists and engineers worldwide. The project exemplifies how advanced analytics can solve real-world problems with global implications, blending technical excellence with environmental stewardship.
Google’s carbon-aware computing system uses predictive analytics to shift workloads across time and location to align with cleaner energy sources, reducing carbon emissions without sacrificing performance.
The system relies on day-ahead forecasting and real-time feedback to guide schedulers, enabling efficient load shaping while maintaining service quality and grid stability.
By moving compute to lower-carbon regions and more energy-efficient clusters, Google has achieved significant reductions in both energy consumption and carbon footprint.
The project has evolved beyond carbon reduction to include peak load shaving, network efficiency improvements, and even emergency grid support during outages.
Google is sharing research, models, and signals publicly to help other organizations replicate and advance sustainable computing practices.
…and 3 more takeaways available in PodZeus
The Journey of a Transformative Idea
“When you have a great idea that affects everybody in the world, is just right and the right team, nothing is actually hard.”
The Challenge of Global Data Center Energy Management
The team discusses the immense complexity of managing energy use across Google’s global data center network, including technical, organizational, and data challenges in shifting workloads sustainably.
How the Carbon-Aware System Works
“We build analytics that understands what's going to happen on the grid... and provide shaping guidelines for shifting work.”
Visualizing Impact and Proving Value
“While the sum of all the changes in compute load add up to zero, the sum of all the changes in carbon footprint add up to something negative.”
Beyond Carbon: Broader Operational Benefits
The system’s impact extends beyond sustainability to include peak load shaving, cost savings, improved network efficiency, and real-world grid support during emergencies.
“When you have a great idea that affects everybody in the world, is just right and the right team, nothing is actually hard.”
“While the sum of all the changes in compute load add up to zero, the sum of all the changes in carbon footprint add up to something negative.”
“We build analytics that understands what's going to happen on the grid... and provide shaping guidelines for shifting work.”
Host
Guests
organization
Anna Radovanovic
person
Valfredo Sirni
person
Ross Conningstein
person
Adrian
person
Franz Edelman Award
other
Resoundingly Human
media
Flex
other
InForms Analytics Plus Conference
other
CO2 Information Grid Modeling
other
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Watt's the Best Time to Compute? Google's Road to the Edelman Award” inside PodZeus.
Start discovering podcast insights today
Start with a 7-day trial and explore a growing catalog of popular podcasts. No credit card required.
No credit card required • 7-day trial • Cancel anytime