You're right, it's the same synchronisation/consistency problem at a different scale, which I honestly find pretty mindblowing. However there is one major diffe

You're right, it's the same synchronisation/consistency problem at a different scale, which I honestly find pretty mindblowing. However there is one major difference. When you add fault tolerance requirements (which you typically do in distributed databases and web applications, but don't always in multithreaded applications) then it changes the approaches and algorithms.