2024.01.16

[seanmcilroy29]

---

2024.01.16 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

  Approved

• Approve previous Meeting Minutes

Approved

Discussion - Adrian

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

Open Issues Project Board

• Cloud provider Power Purchase Agreement (PPA) generation locations as an Impact Framework dataset #20
• Windows Energy Estimation Tool #19
• Carbon Free Energy estimates as an Impact Framework dataset #14 - Comments recently added

Summary Carbon-free energy impact framework Review

Carbon footprint calculation for cloud providers.

• Adrian discusses using plugins to calculate carbon footprint, including the Carbon Free Energy plugin.
• Adrian and Henry discuss the challenges of selecting a data provider for a project, with Adrian suggesting using market-based data and Henry expressing concerns about complying with standards.
• Henry questions the decision-making process and the potential for bias in selecting a provider, while Adrian emphasizes the importance of documenting the calculation methodology.
• Adrian and Henry discuss the impact framework, which allows users to call specific columns for different cloud providers, including Google.
• They consider how to handle location-based data and the lack of information on energy generation through PPAs and other means.

Carbon emissions and energy consumption in data centers.

• Chris Adams suggests that the scope of the analysis should include on-site generation but notes that it is unlikely to have a significant impact on the figures.
• Adrian agrees that rooftop or local generation is small compared to the overall consumption of data centres but raises the question of whether to include provider investments in green energy.
• Chris Adams mentions that the data in the spreadsheet is from a benchmark that was run at some point in the past, and they want to know the difference year on year.

Adrian explains that Google is producing a grade carbon intensity value, a weighted hourly calculation based on electricity maps hourly data, and they are trying to encourage optimisations.

Carbon emissions and offsets for cloud computing.

• Adrian explains the difference between annual and hourly carbon intensity numbers, focusing on Amazon's cloud.
• Adrian suggests including marginal numbers for each region to provide easy reference for cloud instance emissions.
• Adrian explains how Google and AWS approach carbon neutrality through different methods.

Carbon emissions and energy usage in the tech industry.

• Henry questions the logic of buying RECs from different countries to offset carbon emissions.
• Discussion around documenting and understanding annual emissions intensity calculated hourly using average.
• Adrian calculated the hourly grid carbon intensity for Finland, which is 112, despite being 97% carbon free.
• Asim asked for clarification on the hourly grid carbon intensity, which differs from the yearly average.

Energy demand and optimization.

• Henry and Adrian discuss the accuracy of annual energy consumption estimates compared to hourly calculations, with potential discrepancies of up to 30-40%.

Carbon emissions and optimization in the cloud.

• Chris Adams expresses confusion over carbon intensity of AI model, seeking clarification from Adrian.
• Adrian and others discuss a proposed metric for evaluating cloud performance, with a focus on load profiles and market-based pricing.

Using cloud provider data for carbon emissions calculation.

• Asim wants to make it easier for people to access and use energy data in a format that's easy to understand and report on.
• Adrian suggests using a cloud provider's annual emissions data to estimate their hourly emissions, but Asim is unsure if this is accurate.
• Adrian and Asim discuss using a dataset to calculate generic carbon intensity for a region, with the goal of feeding the information into an impact framework. They mention the CFA CFP. Annual as a unique and useful dataset for this purpose, but note that other sources may also be available.
• Adrian suggests picking a low-carbon region for cloud backups to minimize carbon footprint.
• Asim agrees and mentions that business incentives may influence region choice.

Carbon emissions data and its usage in impact framework.

• Asim and Adrian discuss the challenges of calculating carbon emissions for cloud computing services.
• Adrian and Henry discuss using Google's carbon emissions data to create a scope 3 market-based emissions inventory for a company.
• Adrian suggests using Google Doc to reflect on and explain the headers for the spreadsheet.

Using JavaScript and impact framework for data analysis.

• Asim suggests using plugins to simplify data analysis with Impact Framework.
• 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

Milestone Objective

• Outline of proposed milestone dates

AOB

• Any topics members would like to submit

Next Meeting

• 30th Jan

Agreed

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]

I will not be able to attend, but if you can record the meeting and publish it that would be great.
thanks in advance

[seanmcilroy29]

Attended

[PindyBhullar]

Attended

[alexander-kroll]

Attended

[adrianco]

attended

[Henry-WattTime]

Attended

[rossf7]

Attended

[seanmcilroy29]

Summary Carbon-free energy impact framework Review

Carbon footprint calculation for cloud providers.

• Adrian discusses using plugins to calculate carbon footprint, including the Carbon Free Energy plugin.
• Adrian and Henry discuss the challenges of selecting a data provider for a project, with Adrian suggesting using market-based data and Henry expressing concerns about complying with standards.
• Henry questions the decision-making process and the potential for bias in selecting a provider, while Adrian emphasizes the importance of documenting the calculation methodology.
• Adrian and Henry discuss the impact framework, which allows users to call specific columns for different cloud providers, including Google.
• They consider how to handle location-based data and the lack of information on energy generation through PPAs and other means.

Carbon emissions and energy consumption in data centers.

• Chris Adams suggests that the scope of the analysis should include on-site generation but notes that it is unlikely to have a significant impact on the figures.
• Adrian agrees that rooftop or local generation is small compared to the overall consumption of data centres but raises the question of whether to include provider investments in green energy.
• Chris Adams mentions that the data in the spreadsheet is from a benchmark that was run at some point in the past, and they want to know the difference year on year.

Adrian explains that Google is producing a grade carbon intensity value, a weighted hourly calculation based on electricity maps hourly data, and they are trying to encourage optimisations.

Carbon emissions and offsets for cloud computing.

• Adrian explains the difference between annual and hourly carbon intensity numbers, focusing on Amazon's cloud.
• Adrian suggests including marginal numbers for each region to provide easy reference for cloud instance emissions.
• Adrian explains how Google and AWS approach carbon neutrality through different methods.

Carbon emissions and energy usage in the tech industry.

• Henry questions the logic of buying RECs from different countries to offset carbon emissions.
• Discussion around documenting and understanding annual emissions intensity calculated hourly using average.
• Adrian calculated the hourly grid carbon intensity for Finland, which is 112, despite being 97% carbon free.
• Asim asked for clarification on the hourly grid carbon intensity, which differs from the yearly average.

Energy demand and optimization.

• Henry and Adrian discuss the accuracy of annual energy consumption estimates compared to hourly calculations, with potential discrepancies of up to 30-40%.

Carbon emissions and optimization in the cloud.

• Chris Adams expresses confusion over carbon intensity of AI model, seeking clarification from Adrian.
• Adrian and others discuss a proposed metric for evaluating cloud performance, with a focus on load profiles and market-based pricing.

Using cloud provider data for carbon emissions calculation.

• Asim wants to make it easier for people to access and use energy data in a format that's easy to understand and report on.
• Adrian suggests using a cloud provider's annual emissions data to estimate their hourly emissions, but Asim is unsure if this is accurate.
• Adrian and Asim discuss using a dataset to calculate generic carbon intensity for a region, with the goal of feeding the information into an impact framework. They mention the CFA CFP. Annual as a unique and useful dataset for this purpose, but note that other sources may also be available.
• Adrian suggests picking a low-carbon region for cloud backups to minimize carbon footprint.
• Asim agrees and mentions that business incentives may influence region choice.

Carbon emissions data and its usage in impact framework.

• Asim and Adrian discuss the challenges of calculating carbon emissions for cloud computing services.
• Adrian and Henry discuss using Google's carbon emissions data to create a scope 3 market-based emissions inventory for a company.
• Adrian suggests using Google Doc to reflect on and explain the headers for the spreadsheet.

Using JavaScript and impact framework for data analysis.

• Asim suggests using plugins to simplify data analysis with Impact Framework.