What is the difference regarding memory allocation and efficiency between using a struct with two fields and a pair?
As stated in std::pair vs struct with two int's a struct will probably be a little bit faster because no initialization is made.