I recently finished the code for my thesis and wanted to share with you all :). The goal of the thesis was to evaluate language features of Python that were hypothesized to cause performance issues. Quantifying the cost of these features could be valuable to language designers moving forward. Some interesting results were observed when implementing compiler optimizations for Python. An average speedup of 51% was achieved across a number of benchmarks. The thesis paper is linked on the GitHub repo, I encourage you to read it!<p>This was also my first experience with Rust. The Rust community is absolutely fantastic and the documentation is great. I had very little trouble with the "learning curve hell" that I hear associated with the language. It was definitely a great choice for this work.<p>I also included PyPy in my validation section and "WOW". It blew both Cannoli and CPython out of the water in performance. The work they're doing is very interesting and it definitely showed on the benchmarks I worked with.