Not necessarily. I'd check with the application developers first to see if they've encountered the same issue and have a fix or workaround for it. If it's a common application in the field, you shouldn't be the only one to be experiencing the issue (regardless of platform). It could be something that just needs a bit of tweaking (eg, the resource reservations I mentioned earlier). If they haven't encountered the issue before, then it brings it up on their radar and they might research into what's causing the issue. I brought up the SQL issue because what you describe (more resources being consumed and not released until a reboot is performed) is exactly what I've encountered in previous cases. The issue might not be exactly the same, but it could be similar. If the issue is happening in your Windows VM and under MacOS, then it's likely not an OS issue so it's higher up in the structure (so application or data layers). Getting new hardware isn't likely to resolve the issue if it is one of the two layers above. It might allow you to operate longer, but it doesn't address it. I'm not saying don't get new hardware (we all love our toys) but it's not necessarily going to immediately resolve the problem.

One other thing. Reboots might seem to be a hassle (trust me, I hear complaints about it all the time...we see similar threads in this sub as well). But they're necessary (even if the issue is resolved on your end for example). They refresh system resources and also allow security updates to critical files that cannot be updated while the system is up. Doesn't matter what OS you're using. (another debate for another time). At least it's just one box. When you hit enterprise clusters, it's three or more boxes and you have to be very careful, otherwise you can corrupt data very badly.