Join Tables and Rank Users

You are given two analytics tables: - `users(user_id, country, signup_date)` - `listening_events(event_id, user_id, played_at, minutes_played)` Because the data pipeline may retry writes, `listening_events` can contain duplicate rows with the same `event_id`. Write a SQL query to return, for each country, the top 3 users by total `minutes_played` in the last 30 days. For each returned row, include `country`, `user_id`, `total_minutes`, and the user's rank within the country. Then discuss: 1. How you would deduplicate events before aggregation. 2. How you would optimize the query if `listening_events` is extremely large. 3. When partitioning versus sharding would help.