There's no magic here. Your app will spawn worker processes and threads up to the numbers you've configured in your uwsgi.ini (or in your gunicorn config, apache config, or whatever).
If the amount of traffic exceeds your configured capacity, then those requests wait until your app gets to them. If your app is completely swamped and can't respond then the front-end nginx will return a 503 or 504 error.
If you need to handle more requests, then adjust your configuration until you get the performance characteristics you're looking for.
If you're way under your memory quota then try using more workers and fewer threads for better performance (at the expense of more memory usage). Other than that, it's just tweak-test-repeat.
