There's nothing special about reading a large file in perl (or in C, for that matter). You just open it, and read it a line (or block) at a time:

open(IN, "access_log") || die "Can't open file: $!";
while(<IN>) {
    ... process the line ...
}
close(IN);
[download]

Of course here it's the ...process the line... bit that does all the interesting stuff, like parsing the access_log line to extract the information that interests you.

I haven't checked, but I wouldn't be surprised if there was already a module on CPAN which does access_log parsing.

Michael