Benchmarking code

In this recipe, we will explain how you can benchmark your code and how code benchmarking can be used to improve its efficiency.

 

Getting ready

We will show how to benchmark a simple function that takes a single argument in the form of an integer, n, creates a random 10 x 10 matrix of floats, A, and then calculates the norm of a product of this matrix by a random vector, x. The sampling of x should be performed n times and the result of the function should be an n-element vector of calculated norms.

Before we start, please make sure that you have the BenchmarkTools package installed. If it is missing, then add it using the following commands: using Pkg; Pkg.add("BenchmarkTools").

Note

In the GitHub repository for this recipe, you will find the commands.txt file, which contains the presented sequence of shell and Julia commands.

Now, open your favorite terminal to execute the commands.

How to do it...

We will implement the required functionality using two different approaches in...