I want to read the following file into a hash, minus the comment lines (#):

You should read perlopentut for an intro to interacting with files if you are not familiar with opening and reading files. I would read the file in line by line. I would then use next to skip lines starting with '#'. Finally, I would split on a semicolon with a maximum of two terms, based on what you've posted. See perlretut if you are not familiar with regular expressions - they are a core part of Perl and incredibly powerful once you get the hang of them.

#!/usr/bin/perl
use strict;
use warnings;
use Data::Dumper;

my %data;
while my $line (<DATA>) {
    next if $line =~ /^#/;
    my ($key,$value) = split /:/, $line, 2;
    $data{$key} = $value;
}
print Dumper \%data;
__DATA__
# config file
port: 8888
logfile: /data/log
[download]

how would I reference the port number in the hash one the above file is read in?

The short answer is yes. In general, for this sort of question, you should write up a small test script to test the answer in a script. It is also discussed in perldata, the intro to Perl data types.

As a side note, please wrap all code, input and output in <code> tags so things do not get mangled between posting and display. See Writeup Formatting Tips.