gistfile1.pl

#!/usr/bin/perl
use Benchmark qw(:all);
my @strings = (123456, 1234567890, 1234567890123456);

foreach my $str (@strings) {
    print "\nString of " . length($str) . " characters\n";
    cmpthese(
        1000000,
        {
            'length and substr' => sub {
                my $new_str = $str;
                'X' x (length($new_str) - 4) . substr($new_str, -4);
            },
            'regex w/replace' => sub {
                my $new_str = $str;
                $new_str =~ s/(\d+)(\d{4})/('X' x length $1) . $2/e;
            },
            'regex' => sub {
                my $new_str = $str;
                $new_str =~ s/\d(?=\d{4})/X/g;
            },
        }
    );
}

=results

String of 6 characters
                       Rate  regex w/replace             regex length and substr
regex w/replace    248139/s               --              -67%              -82%
regex              740741/s             199%                --              -46%
length and substr 1369863/s             452%               85%                --

String of 10 characters
                       Rate  regex w/replace             regex length and substr
regex w/replace    251889/s               --              -40%              -81%
regex              418410/s              66%                --              -68%
length and substr 1298701/s             416%              210%                --

String of 20 characters
                       Rate  regex w/replace             regex length and substr
regex w/replace    207469/s               --              -24%              -84%
regex              273224/s              32%                --              -79%
length and substr 1282051/s             518%              369%                --

=cut