Yeah I think the key to keep in mind here is that there are two pieces of code present:
- RStudio Server Pro's codebase, which does things like launch R sessions, respond to your clicks in the browser, etc. Some of this is written in C++, but this is just an implementation detail on our part
- The code that you write or use inside of R, Python, etc.
Typically "code performance" is something that you care more about the latter. Even if it were possible to do code performance monitoring on the former, that would just be data you'd have to give us to improve the RStudio codebase (which, in the case of the Pro stack, is not open source).
As @nirgrahamuk articulated, profvis is a very good tool for code performance analysis. If you're looking for more monitoring / logging, you definitely could use DataDog or some other tool for performance monitoring. I'm not sure whether APM would be the right tool for injecting into your code (it may not make sense if your data scientists are not writing C++), but I suspect there are other options.
For RStudio Server Pro monitoring on the systems side, typically what you are looking for is more "overall service metrics and monitoring." That can be done with RStudio Server Pro, although it takes a bit more work to set up today than we would prefer long term 