Bounded sources process data that arrives all at once, while unbounded sources handle data that arrives continuously over time.
Consider whether each source type can be completed in finite time or if data keeps flowing indefinitely.
Think about batch processing for static datasets versus real-time ingestion of ongoing data streams.
Think about what "bounded" implies -- a dataset with a known, finite size versus one that continuously grows.
Consider which data processing paradigms deal with finite datasets and which handle continuous data.
Recall the two main modes of data processing: one for historical data and one for real-time data.

BigQueryIO