Python performance: it’s not just the interpreter
I have a particular view of Python performance from my experience on the Pyston project, and since this view is somewhat nonstandard I wanted to take some time to explain it and give a motivating example.
A common explanation for Python's low performance is that it is an interpreted language. In this post I hope to show that while the interpreter adds overhead, it is not the dominant factor for even a small microbenchmark. Instead we'll see that dynamic features -- particularly, features inside the runtime -- are to blame.
Source: Python performance: it’s not just the interpreter, an article by Kevin Modzelewski.