What is this for (what is your ultimate goal -- Heard of Plucene? Heard of DB_File {would help with memory overhead and the sorting}))?

Now looking at the code...

• One easy way to speed this up is to implement caching. If the timestamp hasn't changed, no need to rescan the entire file/directory.
• Don't build a giant array only so you can build a giant hash afterwards, just build the giant hash ($wordlist{$word}++).
• Why are you keeping a @DirList? You don't appear to be doing anything with it, might think about timestamps again.
• You can write foreach my $SUBDirItem( @SUBDirList ){ ... }
• Are you sure you wanna die (ex, you've reached the last two files, and you can't read the one before last, why not just move on to the next one)?