Given that there are not many collissions hashmaps will give you o(1) performance (with a lot of colissions this can degrade to potentially O(n) where N is the number of entries (colissions) in any single bucket). TreeMaps on the other hand are used if you want to have some sort of balanced tree structure which yields O(logN) retrieval. So it really depends on your particular use-case. But if you just want to access elements, irrespective of their order use HashMap
