You cannot have duplicate keys with a hash. Besides the hard-coded 'data' hash key does not seem to make much sense, and I am not convinced that the data structure you are using is the best one, but it may be because you showed only a small part of your code. If you really need an array of hashes, then you should probably change it to an array of hashes of arrays, or you could possibly use an array of arrays. An example of one possibility under the Perl debugger:

  DB<4> @values = qw / 60 811/;

  DB<5> push @{$AoA[$values[0]]}, $values[1];

  DB<6> @values = qw / 60 812/;

  DB<7> push @{$AoA[$values[0]]}, $values[1];

  DB<8> x \@AoA

0  ARRAY(0x80430518)
   0  empty slot
   1  empty slot
   2  empty slot
   3  empty slot
(...)
   58  empty slot
   59  empty slot
   60  ARRAY(0x803f50f0)
      0  811
      1  812
[download]

This shows, however, that using an array as the first level of the data structure is probably not the best idea: these 60 empty slots are a waste of memory. A HoA might be better:

  DB<1> ($key,$val) = qw / 60 811/;

  DB<2> push @{$HoA{$key}}, $val;

  DB<3> ($key,$val) = qw / 60 812/;

  DB<4> push @{$HoA{$key}}, $val;

  DB<5> x \%HoA
0  HASH(0x80359cb0)
   60 => ARRAY(0x8035ffa8)
      0  811
      1  812
  DB<6> print "@{$HoA{60}}"
811 812
[download]