TLDR; Gröger & Smolinski - Methods to determine the environmental footprint of software – on the way to standardization

Holger presents the Bosch SW Eco Footprint model.

  • Uses analogies of classical LCA stages but separates software inside product and supporting software like cloud services
  • Example of two devices: Coffee Machine & Vehicle Computer
  • Emissions are mapped into Software Stages: Software in Product, Supporting Software, Software Development, Related to software (catch-all category)
  • Each category is further split up into Embodied and Operational carbon emissions showing that they can strongly diverge

Afterwards Jens Gröger presents and addon to the Bosch model

  • When using shared infrastructure (especially true for development or supporting software phase) emissions must be allocated from the total machine to the actual used components
  • Gröger introduces compute primitives:
    • Computing power (compute) (e.g. server, user device)
    • Memory performance (memorize) (e.g. RAM)
    • Storage capacity (store) (e.g. storage system,HDD, SSD)
    • Data transfer (transfer) (e.g. switch, router, mobile communications)
  • Approach works with any environmental impact category (e.g. GWP, ADP, Water, WEEE, TOX)
  • This allows for granular carbon accounting even in highly complex and shared infrastructures

If you liked the Sneak Peak you can find the full slides on the ecoCompute website: https://www.eco-compute.io/files/slides_2025/01_Thursday/02_DIM/05_Smolinski_Groeger_Methods_to_determine_the_environmental_footprint_of_software.pdf