Whether you have already been entered into the scheme, or you have registered and are on the waiting list (we are building capacity as fast as we can!), you are welcome to start working on the items below. And this way, you will be able to get the certification swiftly once we get to you.
All the material related to Green DiSC is covered by a CC-BY-NC-ND 4.0 licence.
For research groups:
General | |
Nomination of a sustainability representative | There is at least one person nominated to lead on sustainability in the group. This role could be called "Sustainability Representative" for example. Should they leave, they must pass on responsibilities. |
Computing sustainability is part of the induction procedure | The group has an induction procedure in place for all new arrivals, explaining the sustainable practices to take. |
Information and resources are shared with other groups engaging with sustainability. | The information collected that could be useful to other groups in the institution (e.g. data centres specs) are shared with a central team and/or other groups engaging with this framework. Material created or improved (e.g. induction or training) is being shared with other groups on the GACert github. |
Dissemination of this program in the institution | The group has engaged across the institution to encourage other groups to think about the sustainability of their work. |
Offices | |
Inventory of office computing hardware | An inventory of the computing hardware used by the lab has been compiled and is updated yearly. This should include: laptops, desktop computers, monitors, external hard drives, accessories (keyboards, mouses, speakers etc.). This should not include HPC equipment (either in a data centre or on-site), as this is a separate criteria. |
Identification of electronic waste processing streams | What happens to electronic waste (e.g. unused computers) is identified. |
Identification of unused computing equipment | Based on the inventory, unused computing equipment has been identified. |
Data Storage | |
Inventory of main data resources | An inventory of all the main data resources is made and updated regularly (frequency depends on field, at least yearly). Details of where this data is being stored are also recorded. Datasets can be considered important because of the value of the data stored (e.g. difficult to replace if lost) or because of the size of the dataset. All datasets used (not only managed) by the group that satisfy one of these criteria should be listed. |
Implement regular cleaning of users own data directories | Users' project directories need to be regularly (monthly is generally a good frequency) cleaned from temporary/outdated files. A process has been implemented to this effect, with regular reminders and a half-way checkpoint after 6 months. NB: this refers to data directory managed by individual team members for their own projects. |
Compute | |
Inventory of HPC infrastructures | An inventory of the HPC infrastructures used by the group has been made. The list covers on-site infrastructures (e.g. server in the office) as well as off-site infrastructures (institutional computing facility and commercial cloud providers). The electricity source of these infrastructures should also be listed (either main power grid, most common, or dedicated facility such as a solar farm). |
Inventory of the most demanding computing pipelines | Each team member lists the most demanding computing pipelines ran (if any), either because of long runtimes or large number of resources, or because the same pipeline is ran a large number of times. |
Training on (green) computing best practices is provided | The group educates team members about green computing best practices. E.g. reducing re-run analysis when not necessary, how to debug on small scale and use checkpointing to limit unnecessary runs, and known coding optimisation best practices relevant to the group's activities. Tips specific to this group are recorded in a document to be updated and improved over the years. |
Others | |
Anything else that you've done? | |
For central teams:
General | |
Nomination of a DiSC representative | There is a point of contact in the central team of the institution which is in charge of coordinating the sustainability initiatives around digital and computational science across the institution. |
Computing sustainability is encouraged to be included in induction procedures | There is material ready around sustainability that can be offered to new starters, and recruiters are encouraged to use it. |
There is a central institutional repository for sustainability resources. | The information collected that could be useful to other groups in the institution (e.g. data centres specs) are collated and made available by a central team. |
Dissemination of this program in the institution | The central team is well placed to encourage individual research groups to engage with this. |
Offices | |
Inventory of the different purchasing streams for computing hardware | A list of the different ways staff acquire computing hardware, and where these come from. This should include: laptops, desktop computers, monitors, external hard drives, accessories (keyboards, mouses, speakers etc.). Include warranty in place for each. This should not include HPC equipment (either in a data centre or on-site), as this is a separate criteria. |
Identification of electronic waste processing streams | What happens to electronic waste (e.g. unused computers) is identified. |
Data Storage | |
Inventory of main data storage | Identifying which infrastructure are user's responsibility and which are centrally managed |
Regular cleaning of centrally-managed data storage | There is a process in place, dates and support for regular clean up of centrally-managed data storage. Research groups are being told how their storage space is being used. |
Template and training resources for data management are made available. | Suitable training for data management plans is key to sustainable data storage. |
Compute | |
Inventory of HPC infrastructures | An inventory of the HPC infrastructures available across the institution has been made. As much as possible, the list covers on-site infrastructures (e.g. server in the offices if possible) as well as off-site infrastructures (institutional computing facility and commercial cloud providers). The electricity source of these infrastructures should also be listed (either main power grid, most common, or dedicated facility such as a solar farm). |
Training on green computing best practices is offered centrally | The institution offers workshops to educate researchers about green computing best practices. E.g. reducing re-run analysis when not necessary, how to debug on small scale and use checkpointing to limit unnecessary runs, and known coding optimisation best practices relevant to the group's activities. Tips specific to this group are recorded in a document to be updated and improved over the years. |
Others | |
Anything else that you've done? | |