being curious about AnoMonk's unpack suggestion, I hacked the following proof of concept.

(NB: you might need to trim the fields!)

use warnings;
use strict;

use Data::Dump;

my %HoH;
while ( my $line = <DATA> ){
    my %fields;
    my ( $slot, @values ) =  unpack( 'A3A17A16A19', $line );
    @fields{'Description','HW Version','Software Version'} = @values;
    $HoH{$slot}= \%fields;
}

dd \%HoH;

__DATA__
 8 CAM (0D, 12U)    CAM-01122W/K05  CMTS_V07.03.01.39  112 days  2:26:
+10
 9 CAM (0D, 12U)    CAM-01122W/K05  CMTS_V07.03.01.39  112 days  2:26:
+10
10 CAM (16D, 0U)    CAM-20016W/G04  CMTS_V07.03.01.39  112 days  2:29:
+46
11 CAM (16D, 0U)    CAM-20016W/G04  CMTS_V07.03.01.39  112 days  2:30:
+15
[download]

OUTPUT:

{
  " 8" => {
        Description        => "CAM (0D, 12U)",
        "HW Version"       => "CAM-01122W/K05",
        "Software Version" => "CMTS_V07.03.01.39",
      },
  " 9" => {
        Description        => "CAM (0D, 12U)",
        "HW Version"       => "CAM-01122W/K05",
        "Software Version" => "CMTS_V07.03.01.39",
      },
  10 => {
        Description        => "CAM (16D, 0U)",
        "HW Version"       => "CAM-20016W/G04",
        "Software Version" => "CMTS_V07.03.01.39",
      },
  11 => {
        Description        => "CAM (16D, 0U)",
        "HW Version"       => "CAM-20016W/G04",
        "Software Version" => "CMTS_V07.03.01.39",
      },
}
[download]

edit

introduced temporary @values to "ungolf" code

update

if you have control over the format, you should consider using a good delimiter like "\t;" ¹ instead of fixed length. Otherwise you'll need to fix code at several places each time you a field-length changes. At least the headlines should be aligned and without whitespace to help scanning field width.

¹) or multiple whitespace /\s{3,}/