For this example, I used your two lines of input data pasted three times in the data.txt input test file:
Python:
import re
data = {}
regex = re.compile(r'CustomerId:(\d+).*?Size:(\d+)');
with open('data.txt') as fh:
    for line in fh:
        m = regex.search(line)
        if (m.group(1) and m.group(2)):
            cust = m.group(1)
            size = m.group(2)
            try:
                data[cust] += int(size) 
            except KeyError:
                data[cust] = int(size)
print(str(data))
Output:
{'1234': 16296, '1237': 16296}
Perl:
use warnings;
use strict;
use Data::Dumper;
open my $fh, '<', 'data.txt' or die $!;
my %data;
while (my $line = <$fh>){
    if (my ($cust, $size) = $line =~ /CustomerId:(\d+).*?Size:(\d+)/){
        $data{$cust} += $size;
    }
}
print Dumper \%data;
Output:
$VAR1 = {
      '1234' => 16296,
      '1237' => 16296
};
solved regular expression to extract only Customerid and Data (bytes) and save in list?