Writing high performance F# code
While this post is addressed to F# .NET developers, it introduces much wider concepts starting from hardware architecture to overall .NET runtime and JIT compiler optimizations. It shouldn't be a surprise - optimizing the application performance requires us to understand the relationships between our high level code and what actually happens on the hardware.
There's a popular opinion that F# code must be slower than equivalent C# code. This opinion is mostly false, however it comes with some rationale. Usually comparison doesn't use equivalent code in both languages, and F# is generally more high level and declarative in nature. "Idiomatic" F# code doesn't always play well with .NET virtual machine. Writing code that is high level, declarative and fast on .NET platform is not an easy task.
A correct profiling of the executing application is crucial before starting any work on optimizing the code - there's no sense in shaving last possible CPU cycles out of the function that's executed for 0.1% of the time.
