Well the problem is a bit more complicated. The solution McA has given certainly works.

Lets say in the loop, I first grep it for a particular user say, stephen by the following expression ,although the later part of this expression needs to be corrected for the server name and etc

if(/^\s+stephen(.*)/)

And , now for the subsequent iteration, I want to grep all the user names and servername,version name for other users who are not stephen. That is becoming a problem to uniquely identify the lines for which the user name is not known. So using split is a problem. Can't I have a regular expression of the following pattern :
$line =~ /^\s+(any user but stephen) (machine name) (version)(server)(date time )/
Hope I could explain the problem.
Thanks.

while(<DATA>){
next if /^\s*stephen\s/;
if (~m/^(\w+?)\s+(\w+?)\s+\((\w+?)\)\s+\((\w+?)\),\s+.+$/){
   print "user: $1 version: $3 server: $4";
   }
}
[download]

(More complicated and less elegant that the use of split in any case IMHO...)