Try to get your data into a convenient structure - i'd suggest an array-of-hashrefs.

Your results should be a 2-D array.

Here is some code to get you started (populated with sample data). You need to figure out how to get your data into this structure:

use strict;
use warnings;

my @points = ( {X=>79.620, Y=>66, Z=>12.3},
               {X=>66.4,   Y=>77, Z=>43.1},
               {X=>73.6,   Y=>55, Z=>93.7} );

my @results;

for my $x(0..$#points){
   my $p1 = $points[$x];
   for my $y (0..$#points){
        my $p2 = $points[$y];
        my $dist = ( $p1->{X} - $p2->{X} ) **2
                 + ( $p1->{Y} - $p2->{Y} ) **2
                 + ( $p1->{Z} - $p2->{Z} ) **2 ;  
        $results[$x][$y] = $dist;
   }
}

#-------------------

for my $r (@results){
   printf "%3.2f\t",$_ for @$r;
   print "\n";
}
[download]

Yes - there is room for optimization, since distances are symmetric, but, at this point, operational and simple code trumps efficient code that introduces complexity.