2024.02.13

[seanmcilroy29]

---

2024.02.13 Agenda/Minutes

---

Time 0800 / 1600 (BST) - See the time in your timezone

• Chair – Adrian Cockcroft
• Chair - Pindy Bhullar (UBS)
• Convener – Sean Mcilroy (Linux Foundation)

Antitrust Policy

Joint Development Foundation meetings may involve participation by industry competitors, and the Joint Development Foundation intends to conduct all of its activities in accordance with applicable antitrust and competition laws. It is, therefore, extremely important that attendees adhere to meeting agendas and be aware of and not participate in any activities that are prohibited under applicable US state, federal or foreign antitrust and competition laws.

If you have questions about these matters, please contact your company counsel or counsel to the Joint Development Foundation, DLA Piper.

Recordings

WG agreed to record all Meetings. This meeting recording will be available until the next scheduled meeting.

Roll Call

Please add 'Attended' to this issue during the meeting to denote attendance.

Any untracked attendees will be added by the GSF team below:

• Full Name, Affiliation, (optional) GitHub username

Agenda

• Approve agenda
• Approve previous Meeting Minutes

Renewable Energy Percentage

• Defining language for the column headers

Action Item: Members requested to assist in defining column headers GSF Real-Time Cloud Renewable Energy Percentage spreadsheet

Open Issues Project Board

• Cloud provider Power Purchase Agreement (PPA) generation locations as an Impact Framework dataset #20
• Windows Energy Estimation Tool #19
• What are the use cases for real time power and carbon intensity information? #12
• How can cloud regions be mapped to electricity grids for carbon metrics? #7
• How can energy data collection and reporting be implemented by a cloud provider? #4
• What should the real time cloud standard be called? #2

Discussion Board

• Review miro dataflow in detail
  • Continue to review and approve the proposed Scope.

Milestone Objective

• Outline of proposed milestone dates

AOB

• Any topics members would like to submit

Next Meeting

• 27th Feb

Adjourn

• Motion to adjourn

---

Meeting Action Items / Standing Agenda / Future Agenda submissions / Links

• The current sheet of raw data is here - this contains guesses for current ranges and is a work in progress at this point
  https://docs.google.com/spreadsheets/d/1RKjD4CuI5bd7JTj-9Mi1-ZhTIc6OW7TH9hUvLPbLsPA/edit?usp=sharing

[nttDamien]

@seanmcilroy29 @adrianco @PindyBhullar
Hopefully you see my comment before the meeting start, if so, it would be great if the meeting is recorded and shared as agreed. It's been several meetings in a row for which the recording was not published.
Have a productive meeting !

[seanmcilroy29]

@nttDamien - The 02 Jan meeting was cancelled due to only a couple people attending. The 16 Jan recording and today's recording can be found here

[seanmcilroy29]

attended

[PindyBhullar]

Attended

[seanmcilroy29]

Summary Notes

AI impact on climate and green initiatives in the tech industry.
Adrian is working on a deep dive into the workload and observability tools for measuring the impact of AI on the Sei.
Sean and Pindy are discussing the potential for green AI and its impact on the Sei, and Adrian's insights will be relevant to the conversation.

Cloud computing carbon footprint and data standards.
Adrian is working on a project to make cloud provider data more useful for modelling and benchmarking, focusing on end-to-end flows and real-time data. The project uses work from the Cloud Native Computing Foundation (CNCF) and Keppler to allocate attribution for workloads in Kubernetes clusters based on CPU utilisation. Adrian and Vincent are working on a cloud carbon footprint tool that provides location-specific data on carbon intensity, carbon-free energy, and other metrics for various cloud providers. The tool aims to close the gap between billing records and cloud providers' data, and it's being developed through a public, open-source approach with input from various stakeholders.

Defining scope and units for energy data.
The group discussed challenges with defining units and naming schemes for data exported from Google's data set. Adrian and Pindy discuss the use of "region" vs "country" in a cloud provider's scope, with Adrian explaining the importance of understanding the scope for energy generation and use.

Carbon emissions data for Google, AWS, and Azure.
Adrian explains using the Impact Framework to retrieve specific data from a large dataset.
Adrian: Google publishes hourly carbon-free energy data for specific regions, while AWS publishes annual data with 100% offsetting in select regions.

Cloud providers' carbon footprint data.
Vincent clarifies AWS's carbon footprint claims, citing hydro and nuclear energy as included in their "carbon-free" label.
Adrian and Vincent discuss cloud provider data, definitions, and homogenization. Adrian hopes to influence AWS's data release process by finding someone still at the company to talk to. Adrian explains that AWS publishes a percentage number (31%) for its carbon-free energy, while Google publishes a numeric proportion (100%) for the same metric. Adrian explains the different types of data Google provides, including annual and location-based numbers.

Location-based vs. market-based carbon emissions calculations.
Electricity maps provide location-based data, but accuracy can vary depending on methodology. Adrian discusses the challenges of calculating Sei in cloud computing, including using different numbers from various vendors.

Carbon emissions data for cloud computing.
Customers want to align with data sources generally accepted by auditors for reporting but optimize for carbon-free only in scenarios. Adrian seeks a clear definition of data for cloud optimization and user complaints to providers.
Adrian and others discuss the importance of accurately defining and labelling data sources in a spreadsheet to avoid confusion and ensure usefulness. The group debates whether to use a separate column for each cloud provider's data, as the data is calculated differently and may require different labels.

Carbon emissions data for cloud computing.
Adrian suggests using location-based emissions factors to make apples-to-apples comparisons between cloud providers.
Adrian explains how AWS and Azure are approaching net zero goals, with a focus on location-based and market-based offsets. Adrian notes that data used for SEI calculations is often old and unreliable, leading to inaccurate decisions about the future. Adrian and Cooper discuss defining headers and filling in data for an impact framework model. The model will pull data from a Google Doc spreadsheet and have headings with documentation for column meanings.

[NAMRATA-WOKE]

@seanmcilroy29 these summary notes are so helpful, thank you! What were the action items that came out of this discussion to support the project?

[adrianco]

Attended