Estimating emissions from ARCHER2

In this blog post we report on work we have been doing to estimate greenhouse gas (GHG) emissions associated with the ARCHER2 service, describe the methodology we use to estimate emissions from the service and the new tools we have developed to help users estimate the emissions from their use of ARCHER2.

The rest of the post is structured as follows: first, we provide a brief overview of sources of emissions and how they relate to ARCHER2, next we look at what how services such as ARCHER2 potentially have a positive impact on emissions. After this we move on to discuss how we have estimated emissions for ARCHER2 and describe tools we have developed to help users (and prospective users) estimate emissions from their use of the service. Finally, we cover briefly future work we have planned in this area.

Categories of GHG emissions

Emissions are typically split into three categories. Only Scope 2 and Scope 3 emissions are potentially relevant for the ARCHER2 service:

• Scope 1: Direct emissions from the service, such as on-site fuel combustion or fleet vehicles. There is nothing of this type currently associated with the ARCHER2 service.
• Scope 2: Indirect emissions related to emission generation of purchased energy, such as heat and electricity. On ARCHER2 this would correspond to any emissions from the electricity used by the service. As we will see, the Scope 2 emissions are reported as zero because the energy contract for the service is based on 100% renewable energy.
• Scope 3: Other indirect emissions from the service. For ARCHER2, this corresponds to the embodied emissions of the hardware and infrastructure from the service, i.e. the emissions from manufacturing, shipping and decommissioning ARCHER2 hardware and supporting physical infrastructure.

The wording above was inspired by the wording used in the Green Software Practitioner course

Impact on reducing emissions

As well as a producer of GHG emissions, HPC systems like ARCHER2 also contribute to reducing emissions. The main source of reduced emissions from services such as ARCHER2 is in the research that leads to new technology, policies and approaches to reducing emissions. Some examples include:

• HPC services run the climate models that are used to provide evidence for setting emissions reductions policies and targets across the world.
• Research and modelling on HPC services leads to development of improved zero emission energy generation by, for example, modelling new wind turbine and wind farm designs.
• Modelling to support the development of new energy storage technologies such as improved batteries.

The emissions reductions from such activities are extremely difficult to quantify for a number of reasons so, at the moment, these are not factored in to the emissions estimates for ARCHER2.

As well as the research activities on the service leading to reductions in emissions, there are other activities that HPC services can potentially take. For example:

• Using the waste heat generated by large scale HPC services as a heat source for homes, businesses or farming. For the ACF data centre where ARCHER2 is hosted we are looking for options on how to do this.
• Incorporating environmental and biodiversity improvements into the service. For the ACF data centre (which is in a rural location) we have been working to improve the site biodiversity and improve habitats. Responsible carbon offset schemes could also potentially be used to reduce emissions if they were undertaken as part of the service.

Estimating emissions from ARCHER2

Scope 3 emissions

Scope 3 emissions from the ARCHER2 hardware have been estimated from a subset of the components that are expected to make up the majority of the emissions. Note that there is a large amount of uncertainty for Scope 3 emissions due to lack of high quality Scope 3 emissions data from vendors. In particular, the number used for the compute node emissions is at the high end of estimated values and the actual value could be as much as 15% lower at around 900 kgCO₂e/node.

We then estimate the per-CU (nodeh) Scope 3 emissions by assuming a service lifetime of 6 years and 100% availability:

7,320,000 kgCO2e / (5,860 nodes * 6 years * 365 days * 24 hours) = 0.023 kgCO2e/CU

All the tools described below use a value of 0.023 kgCO₂e/CU for Scope 3 emissions from ARCHER2.

References:

1. IRISCAST Final Report
2. Estimate taken from IBM z16™ multi frame 24-port Ethernet Switch Product Carbon Footprint
3. Tannu and Nair, 2023

Scope 2 emissions

Scope 2 emissions from ARCHER2 are zero as the service is supplied by 100% certified renewable energy. For information purposes we can calculate what the Scope 2 emissions would have been if the energy was not 100% renewable energy using the methodology described below.

We are aware that there is ongoing discussion in the sustainability community about the impact and effectiveness of certified renewable energy contracts that are supplied through UK National Grid connections. We are monitoring these discussions and taking advice from sustainability professionals on how we report and estimate ARCHER2 emissions.

UK National Grid based Scope 2 emissions are calculated using the compute node energy use for particular jobs along with the carbon intensity of the South Scotland region of the UK National Grid at the start time of the job. The carbon intensity is retrieved from the carbonintensity.org.uk web API.

If the energy use of a job is not available (which happens occasionally due to, e.g. counter failures) then the mean per node power draw from 1 Jan 2024 - 30 Jun 2024 on ARCHER2 is used to compute the energy consumption. This corresponds to a value of 0.41 kW per node.

Estimates of power draw of individual components of ARCHER2 suggest that the compute node power draw makes up around 85% of the system power draw so we also add 15% of the measured energy to the value to estimate the contribution from other components.

Overheads from the electrical and cooling plant will vary with outside weather conditions at the data centre but are typically less than 10%. As a conservative estimate in the tools we develop we add 10% of the total energy to get the estimated energy use including plant overheads.

The final process for calculation the energy use of a job is

1. Retrieve the measured job energy from Slurm (or compute energy based on a node power draw of 0.41 kW if the value is not available from Slurm).
2. Add 15% of this value to estimate energy use by other components.
3. Add 10% of the new value to estimate overheads from the plant.

Tools available to users

To help estimate GHG emissions from your use of ARCHER2 and place them in context to other sources of GHG emissions we have used the estimation methodology described above to develop two tools - one to estimate historical emissions from jobs that have already been run and one to estimate future emissions associated with use of ARCHER2.

At the moment, the following tools are available:

• jobemissions - a command line tool on ARCHER2 that reports estimated emissions for a specified, completed job. It can also provide comparisons to other GHG emissions sources.
• ARCHER2 CU Calculator provides estimated emissions for the specified CU use of ARCHER2.

More detailed information on the use and output from these tools is available in the ARCHER2 documentation: Estimating your emissions

Future plans

We are continuing to work on improving the tools available to users and are currently developing a command line tool that will allow projects to estimate their historic emissions associated with their use of ARCHER2 over different periods. We hope to have this new tool available to users soon.

We are also continuing to try and improve the emissions estimates we have based on updated information and new developments in the area of embodied emissions in particular.