My feeling is that you are using a few operations, which are all not very fast. split() allocates a few strings, which takes some time, and add() for sequences takes also some time. Maybe your parseInt() is not really necessary, may a plain string compare work?
Note that parsing files was discussed in the Manning book, and I recently added a section to my book too: <https://ssalewski.de/nimprogramming.html#_parsing_data_files_in_parallel>. And I think there was a blog post called "Fast CVS parsing" or so many years ago, but I was not able to find it again.
