This page is made to gather all the sources used in our GreenIT studies, and is to be used for specific cases, when we encounter a new type of service, or want to study a specific point of a digital service. No template or module is associated to it.

1. What is a digital study?

Scope

1.1 Servers

The servers studies are usually already covered by regular modules (Cloud or on-premise)
In the case you encounter a very specific case that needs a servers study, the Generic cloud template should be used.
This part covers the use of servers for compute and storage services, including the electricity consumption, amortization and gas leaks of electronic devices.
If you have to do a very specific servers study, there are chances that you won't know much about the type of server used. What should be done is using average values provided in the servers tab.
To account for data storage, make sure to use both the electricity and amortization emission factors, and note that the unit is in GB.month (Go.mois in french),
which means you should multiply it by 12 if you only have the amount of data stored in GB. Also, there is a ratio *8 if your collect data in Gb instead of GB.

1.2 Network

The network part represents the electricity consumption and amortization of the network infrastructure, that connects data centers with end-users (or others data centers).
This part can be relatively straightforward: You'll collect or calculate the amount of data transfered, and multiply it by the emission factors. But a few things need to be checked:

Do you include the network amortization? We don't always include it, because it is a very uncertain part (we don't know the distance travelled by the data), but it is required for some frameworks.
What type of network are you using? Is it fixed (Wifi), mobile (5G), satellite? The infrastructure isn't the same at all, and so are the emissions
For servers, it is also different, as the inter-servers network is much more optimized than the regular internet network

This toolbox provides you some ways to estimate the amount of data transfered, depending on the type of content that is shared, in case you cannot measure it directly.

1.3 End-user devices

This section is usually the one not included in our regular studies, because it represent a part that is more difficult to estimate, and a downstream impact.
Most of the custom GreenIT studies revolve around estimating the impact of end-users. This is something included in the "use of sold products"analysis, but you might need to run some custom calculation sometimes
Similarly to other sections, you'll need to calculate the impact of the energy consumption of these devices, and potentially their amortization. The main hypothesis are around the average power, the average use duration and how we amortize these equipment.

This toolbox provides you with normalzied values to estimates these impacts.

2. How can I be sure I'm measuring the right thing?

The first thing is to validate the scope of your study:

What do you need to take into account (for a GHG Assessment, we probably won't need the end-devices amortization, but we'll need their energy consumption)
What is already taken into account somewhere else (financial data, e2e or advanced modules)
Is there a risk to double count things?
Is it worth to spend time measuring that? (e.g: email campaign usually doesn't represent much emissions)