listing unique file prefixes in a directory Linux

Hi guys,
I'm trying to achieve the following.
I have a folder with the following path: /home/service
in this folder I'll have many files all with the same format
user1.node1
user1.node2
user2.node4
user1.node3

and etc and the order can be randomized each time.
each of those files has a simple integer in them. and the prefix (user#) corresponds to a certain in user a DB.
basically what I need to do, is to read the current integer from the DB for the selected user (I already have the command for that, it's tested and working) and then. to add to it ALL the integers of files corresponding to that user and write the new total (meaning current + sum of all integers of said user) back into the DB (The writing is done by calling a python script that accepts the user name and the total as arguments)

how would I go about achieving the following?

The script I have at the moment isn't working so well.

files=$(find /home/service/ -type f |grep -v "bash" | sed 's#.*/##')
for file in $files
do
  	user=$(echo $file | awk -F'[.]' '{print $1}')
        total=0
        curr_traffic=$(echo "select traffic from campaigns.Campaigns where UserName='$user';" | mysql -N -u ##### -p#####)
        let total=$total+$curr_traffic
        echo "pre DB write total is $total"
        ufiles=$(find /home/service/ -type f |grep -v "bash" |sed 's#.*/##' |grep $user)
        for ufile in $ufiles
        do
          	traffic=$(cat /home/service/$ufile)
                let total=$total+$traffic
        done
	echo "Updating the traffic for $user as $total"
        rm -f /home/service/$user.*
       	#./update_statistics.py $user $total
done

Select allOpen in new window