sub read_initial_data
 {
        my $config = Config::Tiny->read('/home/database_update.txt');
        foreach my $section (sort keys %$config)
        {
                print "[$section]\n";
                foreach my $parameter ( keys %{$config->{$section}})
                {
                        $parameter = $config->{$section}->{$parameter}
+;
                        my $sth = $dbh->prepare(qq{update $section set
+ Account_balance=20000 where Id = 101});
                        $sth->execute();
                        $sth->finish();
                        print"$parameter\n";


                }
        }

 }
[download]

[Subscriber]
Id=101
Account_balance=2000000
[download]

"This code is not working at all", "its is not working.". There is little difference between this and Re^2: Using perl to parse a file and update values into a database, where you've already been given the similar feedback (here). Try reading and understanding the responses you already have and the links you've been given.

As far as I understand, the first config element in each section is the identifying key for the update. For example UPDATE .... WHERE id=101. Since you cannot determine the parameter order using  keys %{$config->{$section}}, how will this work predictably ?

Other questions are
1.. Are there updates to many records in a table ?. In other words, does the [Subcriber] section appear many times in the text file ?.
2.. Is the order in which the updates are applied important ?

poj