Real-time ride-hailing system ingests driver locations every 5s and booking events, aggregates supply/demand for the last 30s every 2s, and stores results for low-latency dashboards. Which windowing and storage choice fits this requirement?

A hopping window is ideal here because it allows for overlapping windows, enabling the system to aggregate data for the "last 30 seconds every 2 seconds." This means a 30-second window slides forward by 2 seconds each time, capturing the most recent data for continuous supply/demand calculation. Memorystore (Redis) is chosen for storage due to its low-latency read/write capabilities, which are crucial for real-time dashboards. Tumbling windows are fixed and non-overlapping, so they wouldn't provide the continuous 30-second view. Session windows group events based on user activity, which is not applicable here. BigQuery is an analytical data warehouse optimized for large-scale queries, not for low-latency, real-time dashboard updates.