And, Perl's hash implementation can perform better compared to Awk's associative arrays. The `SCOWL-wl.txt` file used below was created using http://app.aspell.

And, Perl's hash implementation can perform better compared to Awk's associative arrays. The `SCOWL-wl.txt` file used below was created using http://app.aspell.net/create http://app.aspell.net/create. `words.txt` is from `/usr/share/dict/words`. Mawk is usually faster, but GNU Awk does better in this particular case. $ wc -l words.txt SCOWL-wl.txt 99171 words.txt 662349 SCOWL-wl.txt 761520 total # finding common lines between two files # shorter file passed as the first argument here $ time gawk 'NR==FNR{a[$0]; next} $0 in a' words.txt SCOWL-wl.txt > t1 real 0m0.376s $ time perl -ne 'if(!$#ARGV){$h{$_}=1; next} print if exists $h{$_}' words.txt SCOWL-wl.txt > t2 real 0m0.284s