This is a simulation of the famous coupon collector's problem. The problem is as follows:

Given a set of $n$ coupons, each with a unique number, how many coupons do you need to draw with replacement until you have drawn each coupon at least once?

The answer is $nH_n$, where $H_n$ is the $n$ th harmonic number. This simulation will run the experiment for a given number of coupons and write the average number of draws to the standard output.

cargo run

The mathematical simulation lies in the trial function. The function creates an array called collected, which has N elements, each being false. Each element in this array represents whether that element has been collected. While any of the elements are false, it will keep generating a random number between 0 and N and setting the corresponding element in collected to true. Once all elements are true, it will return the number of draws it took to collect all coupons, completing one trial of the simulation.

The rest of the program is just a loop that creates a few threads and runs the trial function in each thread. It then takes the average of the results and prints it to the standard output.