Maybe it could answer the question for an already indexed data set, but it probably couldn't INSERT 1 billion rows and then answer the question within that time

Maybe it could answer the question for an already indexed data set, but it probably couldn't INSERT 1 billion rows and then answer the question within that time. The ingestion is part of the challenge.