I went straight to the user-side numbers first, average emails per person per day, attachment sizes, that kind of thing.
Break down the problem into user behavior, email characteristics, and network overhead. Estimate daily active users, average emails sent/received, and average email size, then calculate total data volume and convert to required bandwidth. Consider Mars-specific constraints like latency and potential offline usage.
Pro tip: Acknowledge that bandwidth is not just about average throughput but also peak usage and latency; propose a tiered or asynchronous sync approach to handle Mars-Earth communication delays.
Estimate the number of active users and their email activity. Assume all 10,000 colonists are users, with varying activity levels (e.g., 80% active daily).
Determine average emails sent/received per user per day and average email size (including attachments). Use typical enterprise email metrics as a baseline.
Multiply users by emails per user by average size to get daily data volume. Convert to required bandwidth (e.g., Mbps) considering peak hours and protocol overhead.
Account for high latency (up to 24 minutes one-way) and potential disruptions. Consider store-and-forward or batch transmission to optimize bandwidth usage.
Sanity-check assumptions against known benchmarks (e.g., Gmail usage on Earth) and propose a pilot or phased rollout to refine estimates.
AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.