Your question is Optimize Pipeline for 10x Throughput. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
Given a fixed budget and hardware constraint, how would you optimize an existing data pipeline to handle 10x the throughput?