Bioinformatics/Computer Labs/Servers

Computers

If not running programs overnight, turn off your computer when leaving the office. Use any power-saving modes available

  • Set your computer to go to sleep when not touched after a certain time frame.

Don’t use screen savers. They do not save much energy because they generally keep the CPU operating. If you need the CPU running (running some modeling for example), then just turn the screen off.

Reduce your computer monitor’s brightness from 100% to 70% saves up to 20% of the monitor’s energy consumption. It can be better for your eyes too.

Emails

Each email sent emits 4g of CO2 to the atmosphere and if the email includes an attachment it emits another 50g of CO2 or more.

Reference: Mike Berners-Lee: ‘ How Bad are Bananas?: The Carbon Footprint of Everything’ (2010)

Unsubscribe from unwanted mailing lists
Unattach attachments in email chains

Video Conferencing

An hour of videoconferencing or streaming, emits 150-1,000 g of carbon dioxide. Compare this to a gallon of gasoline burned from a car emits = 8,887 g) and requires 2-12 L of water.

Leaving camera off during web call can reduce these footprints by 96%

Source: https://www.sciencedaily.com/releases/2021/01/210114134033.htm

Printers

Set sleep modes for printers
Set double sided printing in black and white as the default

Run Smart

Run using the smallest dataset

  • When testing out new algorithms, techniques, or simulations, use the smallest dataset or smallest simulation time frame possible to see if things are working. It will let you see any errors quickly. If working, then scale up. Put in a continue function, so that if it is working, you can append new data to the file and not run these datapoints again.

Elegant programing and use updated versions

  • Often the first version of a software or program is clunky and not optimized. Update software to the newest version - in some cases, it can save you 70% of the energy and shorten run time.
  • If you are programming yourself, look at efficient coding practices.

Extend the life of research data

Well organised, standardized, well documented, preserved and shared data allows others to use your data again in the future. It means fewer experiments to run.

Example: GWAS (Genome-Wide Association Study) - Community-sourced database on the association of specific genetic variations with particular diseases. Scientists can look up values there, rather than re-running.

Other Databases:

Choose open source software

There have been examples where certain software licences expired or the software company no longer supported it. If your data is in a particular format that can only be read by that software, any issues with the licencing or software provider renders your data useless

Servers

  • Using servers for temporary storage/computing
    ** instead of buying your own computer, central/aggregated/cluster/cloud computing means you can rent space for a short period of time and share resources with others

  • Use renewable sources for electricity

  • Consolidate server resources to a centralized location to allow for one larger, more efficient cooling system

  • If possible, redirect waste heat from server rooms to heat other parts of the building

  • Look for environmentally friendly equipment
    ** the manufacture, transport and maintenance matters
    ** Look at software and hardware choices, parallelization, memory allocation, run duration, CPU and GPU types and server location

  • Consider lengthening the time between server unit replacement
    ** Extend the life of server units by prioritizing newer servers for intensive tasks and use older servers for less intensive tasks and storing backups
    ** Store pre-processed data and less important data in older storage units
    ** Instead of storing a 100% Backup on two drives, split over multiple drives so that if one fails, your data is not lost

  • Opt for large data centers
    ** There is a lot of variability between the efficiency of data centers. 40% of the energy usage goes towards coolig. Large data centers tend to be more energy efficient because they have pooled resources, more money for efficient equipment and larger coolers tend to be more energy efficient in general.

  • Decide how important the data is and control how many backup copies to have and where

  • Delete outdated datasets

  • Put data on external hard drives for backup
    ** Could storage requires 1W/GB, but must be on 24/7 to allow the data to be accessible at all times. Hard drive storage means it uses energy only when you need to access your data.

  • When setting up cloud computing, limit tasks to 24 hours of computing time
    ** This forces users to be more conscious by demanding they resubmit every day. This means that no one accidently submits huge tasks without knowing it and wasting resources

  • Study of carbon footprints of bioinformatic processes 2021.03.08.434372v1.full.pdf (627.9 KB)

  • tool to estimate the impact of your own computational work

Carbon Footprint Examples: 100nm Molecular dynamic simulation of virus = 18-95kg CO2 e = 19-105 months of a tree to sequester

End of Life

Keep your devices as long as possible. 70-80% of a consumer device’s carbon footprint is contained in the production. Do what you can to prevent a new machines from being produced

  • upgrade your existing system
  • Get it fixed

If you are looking to get rid of old computers, laptops, printers, harddrives, mobile phones, etc., ask if there’s anyone in your institute who wants them. Companies often replace their technology every couple of years. They might not have the most up to date operating systems, but are still perfectly good for most residential purposes.

Instead of trashing old machines, consider donating them. There are charities that take computers for use in schools, libraries, and disadvantaged homes.

See our Waste reduction - Sharing/Donating/Buying Used/Fixing section

Donate them to organizations such as:
Free Geek - non-profit that diverts E-waste that would otherwise be recycled or thrown away, refurbish it, and give it back to our community at no or low cost. They run free technology fixing clinics and they even have an E-bay store.

  • Operational in US (OR, AK, GA, IL, MI, MN, PA), Norway and Canada (ON and BC)

Recyclers

As with all other lab waste, it’s very important to ensure it is recycled responsibly and not just shipped off to developing countries. Look for recyclers who are certified Basel Action Network e-stewards (covers
Canada, Brazil, Germany, Mexico, Singapore, South Korea, United Kingdom and USA) or operate under the EPRA (Canada).

Heading 1

Heading 2

Heading 3

This article is a stub. You can help by expanding it. Get Involved!

1 Like