fwiw, rindex appears to be slightly faster:

use Benchmark qw(timethese);

my $s = "abcdefg";

timethese( 6_000_000,
    {
        rindex => sub { my $i = rindex( $s, "cde" ) },
         index => sub { my $i =  index( $s, "cde" ) },
    }
);
[download]

Benchmark: timing 6000000 iterations of index, rindex...
index: 1 wallclock secs ( 1.14 usr + 0.00 sys = 1.14 CPU) @ 5263157.89/s (n=6000000)
rindex: 1 wallclock secs ( 1.02 usr + 0.00 sys = 1.02 CPU) @ 5882352.94/s (n=6000000)

their functional difference is that rindex starts from the end of the string, and index starts at the beginning. there is no difference in their usage.