Reconstructed SHELL program:

#!/bin/sh 

TEMPFILE0=/home/h150371a/file.tmp
TEMPFILE1=/home/h150371a/file1.tmp

sudo /usr/openv/netbackup/bin/admincmd/bpimagelist -A -media -hoursago
+ 24 |sed ' s/\|/ /'|awk '{print $1}' > $TEMPFILE0
cat $TEMPFILE0 | while read Image
do 
        sudo /usr/openv/netbackup/bin/admincmd/nbemmcmd -listmedia -me
+diaid $Image | grep -iE 'Media ID|Data Expiration' >> $TEMPFILE1
done
if [ -s "$TEMPFILE1" ]
then 
        mailx -s "Daily" ishonbackup.nar@capgemini.com < $TEMPFILE1
fi 
rm $TEMPFILE1 -MP
# rm $TEMPFILE0
[download]

The only way I see this becoming faster in perl is if you do not write the files to disk, at expense of more RAM usage during execution.

NetWallah already did a good job of presenting you sample code... except you would get multiple emails instead of a single one... might want to replace that.

maybe this (also untested)

my @MEDIA;
for (qx<sudo /usr/openv/netbackup/bin/admincmd/bpimagelist -A -media -
+hoursago 24>){
  s/\|/ /;
  my ($Image) = split;
  for (qx<sudo /usr/openv/netbackup/bin/admincmd/nbemmcmd -listmedia -
+mediaid $Image>){
    next unless m/Media ID|Data Expiration/i;
    push @MEDIA, $_;
  }
}

if(@MEDIA){
  open(FF, "|/usr/bin/mailx -s 'Daily' ishonbackup.nar\@capgemini.com"
+) or die $!;
  for (@MEDIA){
        print FF $_ . "\n";
  }
  close FF;
}
[download]