http://www.perlmonks.org?node_id=481573


in reply to How can I improve the efficiency of this very intensive code?

It may be that sorted arrays are the way to go, as mentioned elsewhere. But it might make sense to go from the data you have now to an inverted list of best matches, so as to sort each list only once:

my $inv = {}; for my $index (keys %$rc) { my $subhash = $rc->{$index}; my @sorted = sort { $subhash->{$a} <=> $subhash->{$b} } keys %$sub +hash; if ($subhash->{$sorted[0]} < $subhash->{$sorted[1]}) { # this is the best match push @{ $inv->{$sorted[0]} }, $index; } }
and now the strong match test becomes:
return (@{ $inv{$id2} } == 1) ? 1 : 0;

I'm not sure that I fully understand the problem, so the code might need some tweaking ...

Hugo