Introducing the Benchmark Package: Complementing Unit Tests with Performance Checks

Hi everyone,

I wrote an introductory blog post to the benchmark package here feel free to discuss here or in the benchmark related project subtopic for concrete questions.




I wonder if you can go into more detail about your experiences with Swift performance. There is a well-known trade-off between usability and performance, with Python choosing one extreme and C++ another, while Swift tries to balance the two better.

Since you are so focused on performance for your trading work, how have you found Swift's particular choices affecting your work? Have you started using the newer features that are more about performance?

Any info about how the language has helped or hindered your efforts would be interesting.

I think I'll revert at a later point outlining that experience, we are indeed very happy with e.g. the memory ownership work in progress (and some of the control we've received for task execution), but as have been outlined in my other post of our experience, there are some pitfalls too (most of which are known pain points, which hopefully will receive some TLC going forward).


Great to see this package being featured on the blog! I used it last year for the first time and was very impressed. Great work @hassila et al!