As a side note, please be aware of SQL_Injection, you have had this pointed out to you a couple of times (Re^4: Problem passing date to SQL, Re^3: After parsing .xls the rows getting emerged). The DBI documentation has a section on using placeholders, the links you've been given previously have further examples.

First, I have to second marto in that you should always use placeholders, third (since second is already used) you could aggregate all your columns into one call:

my @column_names = ("MO_resp","MT_resp","AO_resp");
my $sql = "select ".join ",", map { "sum($_)" } @column_names;
$sql .= " from ? where Date = ? and Hour like ?";
my $hour_sth = $hour_db->prepare( $sql );

$hour_sth->execute($table_name, $db_date, "$hour:%") or die $DBI::errs
+tr;
[download]

See join and map.
If you are iterating over $db_date and $hour, you should keep the call to $hour_db->prepare outside those loops to reduce the overhead of binding placeholders in the SQL statement.

Hey Programatic you are also doing the same but you are passing all the column names once by using join & map. Good thought. But in place holders If i place table name its showing a error.

DBD::mysql::st execute failed: You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near ''CDR' where Date = '13-10-2015' and Hour like '00:%'' at line 1 at DB_Test.pl line 25.

So I replaced my '?' with table name. Then its working good. Even though I used your logic I need to maintain all my column names in array. I want to change that. Here I am writing all columns in program into big arrays which is looking odd. Can I get that coloumn names from db???

Sample (Tested)

#!/usr/bin/perl
use strict;
use warnings;

use DBI;
my $db_date=$ARGV[0]; #04-10-2015

my $hour = "00";
my $db_name   = "ravi";
my $table_name = "CDR";
my $host      = "xxx.xx.x.xxx";

my @CDR_array = ("Total_MO","Total_MO_sucess","MO_userdep_error","MO_s
+ubcrib_error","MO_system_error","Total_MO_success_rate","Total_AO","T
+otal_AO_sucess","AO_userdep_error","AO_subcrib_error","AO_system_erro
+r","Total_AO_success_rate","Total_MT_P2P","Total_MT_sucess_P2P","MT_P
+2P_userdep_error","MT_P2P_subcrib_error","MT_P2P_system_error","Total
+_MT_P2P_success_rate","Total_MT_A2P","Total_MT_sucess_A2P","MT_A2P_us
+erdep_error","MT_A2P_subcrib_error","MT_A2P_system_error","Total_MT_A
+2P_success_rate","Total_AT","Total_AT_sucess","AT_ESMEdep_error","AT_
+network_error","AT_system_error","Total_AT_success_rate","P2P_FDA_cou
+nt","P2P_FDA_success_count","P2P_FDA_success_rate","A2P_FDA_count","A
+2P_FDA_success_count","A2P_FDA_success_rate","Total_Count","Total_TPS
+","Greatest_TPS");

my $dbh = DBI->connect("DBI:mysql:database=$db_name;host=$host;mysql_s
+ocket=/opt/lampstack-5.5.27-0/mysql/tmp/mysql.sock","root","",{'Raise
+Error' => 1});

my $sql = "select ".join ",", map { "sum($_)" } @CDR_array;
$sql .= " from CDR where Date = ? and Hour like ?";

my $sth = $dbh->prepare( $sql ) or die "Cant get prepare";
$sth->execute($db_date, "$hour:%") or die $DBI::errstr;

my @my_array = $sth->fetchrow_array();

print $my_array[1],$/;
print @my_array,$/;
[download]

Can I get that coloumn names from db???

You can get table and column names from the db using queries:

my (@tables, %map);
my $q = $dbh->prepare("show tables");
$q->execute or die "Error listing tables";
        
while ( my ($table) = $q->fetchrow_array ) {
    push @tables, $table;
}
$q->finish;
for my $table (@tables) {
    $q = $dbh->prepare("describe $table");
    $q->execute or die "Error describing table $table";

    while ( my $hr = $q->fetchrow_hashref ) { 
        push @{ $map{$table} }, $hr->{Field};
    }
    $q->finish;
}
[download]

Now you have a hash %map keyed on table names, whose values are anonymous arrays containing column names. Use this array to build your query:

my $table_name = "CDR";
my $sql = "select ".join ",", map { "sum($_)" } @{$map{$table_name}};
$sql .= " from $table_name where Date = ? and Hour like ?";
[download]

perl -le'print map{pack c,($-++?1:13)+ord}split//,ESEL'

My nick feels honoured being linked to something as old as FORTRAN - but the concept is much older; it was invented after speech and before writing, and eventually led to an early cosmic calculator (though at that time writing already had been invented, I guess).

I have heard (through memory sharing) that australian aborigines still have a vast block of shared memory, which they access entirely wireless.

Before the concept of shmem, there was fork.

perl -le'print map{pack c,($-++?1:13)+ord}split//,ESEL'

well, originally I wanted to refer to Wikipedia, but I also wanted the title to be "shmem", and Wikipedia "shmem" is about something more specialized.

Besides, you are much longer with Perlmonks than I, so Fortran seems appropriate :-)

BTW thanks for the hint to the Phaistos Disk. Will read it.