2024.03.26

[seanmcilroy29]

---

2024.03.26 Agenda/Minutes

---

Time 0900 (PT) / 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

• RTC Cloud Region Impact Framework Documentation

Open Issues Project Board

• AWS PUE Data Alignment #29
• GCP PUE Data Alignment #28
• Azure PUE data alignment #27
• 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

• 09th Apr

Adjourn

• Motion to adjourn

---

Previous meeting Action Items

• Write a doc section defining terms like hourly, 24/7, grid, provider, etc.
• Reach out to Microsoft rep (Thomas) to discuss PUE data needs
• Create GitHub issues to request PUE data from AWS and Google
• Reach out to AWS contact to discuss getting power usage metrics
• Write a paragraph in the doc to explain the difference between market-based and location-based carbon intensity metrics
• Add section to doc with table/graph to show dimensions of different carbon metrics (geography, time, production vs consumption, etc)
• Follow up on Google PUE data issue

Members to review and comment on the RTC Cloud Region Impact Framework Documentation
Add a column for Google's 24/7 carbon intensity data once it is approved for publishing.
Members requested to assist in defining column headers GSF Real-Time Cloud Renewable Energy Percentage spreadsheet
Fill in missing data columns in the spreadsheet, including PUE data and IEA annual average numbers.
Get the spreadsheet definitions finalized to start filling in the data. Clarify column names and descriptions.
Review and comment on Scope Flow

[adrianco]

attended

[llhardy1A]

attended

[alexander-kroll]

attended

[seanmcilroy29]

Attended

[seanmcilroy29]

RTC Cloud Region Impact Framework Documentation

[PindyBhullar]

Attended

[seanmcilroy29]

MoM
Adrian opens the meeting at 1000 (PT) / 1600 (BST)

Sustainability and energy efficiency in cloud computing.
Adrian provided a group update on sustainability efforts, noting that while there has not been any immediate progress, there are encouraging signs. Adrian requested that Microsoft provide a tabular breakdown of Power Usage Effectiveness (PUE) data for every region, along with clear documentation of the methodology used. Adrian and Sean discussed energy estimation and geolocation data.

Carbon emissions data and naming conventions.
Adrian has updated the proposal for the impact framework by incorporating geolocation data. During the meeting, Adrian explained how the Impact Framework can be utilized to analyse carbon emissions data for Google and Azure cloud providers. Adrian and Cooper discussed the naming scheme for the emissions factors, specifically focusing on whether to use "hourly" or "24/7" to describe the level of granularity of the data.

Carbon emissions data and reporting methods.
Cooper: Location-based calculation should be reported separately from market-based calculation.
Adrian: Google data may not match SEO needs, and different providers may have different methodologies.
Adrian and the group discuss electricity maps, consumption, and carbon intensity in a CI. Adrian explains the Impact Framework, a tool for analysing data from various sources to understand the time it takes to produce results.

Carbon emissions data for cloud computing.
Adrian provides missing data for Azure regions, including geolocation numbers for cities. Henry suggests using a dash provider to indicate the source of emissions factors, as different providers may use different data sets or calculation methods. Henry suggests using energy information from the centre provider and independent sources to calculate grid carbon intensity. Adrian notes that they need to pull numbers and fill in a spreadsheet with the correct numbers for each region and suggests contacting providers for electricity maps.

Improving data accuracy for cloud computing emissions.
The group discussed data availability and how different release formats affect it. Adrian took an Action Item to create an issue for each cloud provider and requested more information on specific operating systems, asking for progress updates. Adrian also asked for more detailed data on carbon emissions, with hourly data, to move on from a 'stuck' spreadsheet.

Energy usage in cloud computing.
Group discussed the keys for a cloud provider and region, with Adrian highlighting the importance of transparency. Adrian suggests two approaches for measuring power consumption in cloud instances: using Intel metrics and collecting real-time data from providers. Adrian suspects AWS measures energy usage but doesn't expose it, while Google may do the same. Adrian believes CPU utilisation is a standard metric but is not always accurate due to clock rate and thermal behaviour.

Action Items

• Create specific issues for each ask to cloud providers
• Escalate issues to cloud provider contacts to get carbon and energy data
• Follow up in 1-2 weeks on any progress with Microsoft contact
• Look into getting publicly available annual grid carbon intensity data
• Clarify if Google's carbon intensity data is helpful for CI or not
• Check internally on providing GCP PUE estimates
• Figure out naming conventions for metrics like hourly vs 24/7
• Add columns in the spreadsheet for Electricity map and WattTime data
• Fill in missing data in a spreadsheet for regions
• Add geolocation data for regions in the spreadsheet