How do you manage concurrency and asynchronous operations in Node.js when processing large-scale data streams?