I could list some websites where you could announce your game, but the process to find non-paying players is pretty much the same as the one to find paying players. That means I would just be duplicating the answers to "How can I promote my game?" and similar questions.
The problem with public beta tests is that they easily escalate to a PR event for the game, which can be a mistake if the product is not yet polished enough for the public. A professional studio would pay beta testers and have them sign an NDA which forbids them from talking about the game and all of its bugs and flaws, but I have the impression that this is outside of your budget constraints.
When it is just about stress testing your server technology, you might consider writing a headless bot to play your game and start a few hundred instances of that bot. The advantage of this is that you can easily scale the load accurately and also repeat your test with minimal organization overhead when you want to find out how some minor changes affect the overall performance. When you would like to simulate real-world network performance as well, you might want to take a look at the question "How can I simulate a bad internet connection for testing purposes?".
It can sometimes be hard to accurately simulate player behavior, because players sometimes behave in ways which are hard to predict for the developer which might result in far different server load than expected. That's why it is a good idea to have a small circle of testers you trust personally whose behavior you can use as a basis for your bots behavior.