Practice sustainable software architecture vocabulary: choosing low-carbon regions, energy-efficiency trade-offs, reducing unnecessary compute, and eliminating always-on services.
0 / 5 completed
1 / 5
The architecture decision record says: 'Choose the region with the lowest ___ intensity.' What determines this choice?
Carbon intensity of a cloud region is measured in gCO2e/kWh. Regions powered by more renewable energy (e.g., Nordic regions for AWS/Azure/GCP) have lower carbon intensity, meaning the same workload emits less carbon there.
2 / 5
The team documents architecture ___ offs for energy efficiency: some choices that improve performance increase energy use.
Architecture trade-offs for energy efficiency acknowledge that not all optimisations go together. Caching improves speed but adds stateful infrastructure; serverless reduces idle compute but may have cold-start overhead. Sustainable design requires explicit trade-off analysis.
3 / 5
The refactoring initiative claims: 'This design reduces ___ compute by 40%.' What type of compute is being cut?
Unnecessary compute is any processing that consumes energy without delivering value — including over-provisioned always-on servers with low utilisation, busy-waiting loops, redundant data processing, and speculative pre-computation.
4 / 5
The green architecture review notes: 'We eliminated 3 always-___ services.' What were these?
Always-on services are infrastructure components that run continuously regardless of traffic — often virtual machines or containers kept running to avoid cold-start latency. Replacing them with event-driven or serverless alternatives can significantly reduce idle energy consumption.
5 / 5
The team measures ___ before and after the optimisation to quantify the sustainability improvement.
Measuring energy consumption (kWh) before and after an architectural change gives a direct, quantifiable sustainability metric. Combined with the regional carbon intensity, it translates directly into CO2e saved.