Comparison of hash table performance with open addressing and closed addressing: An empirical study

Abstract

In this paper, we conducted empirical experiments to study the performance of hashing with a large set of data and compared the results of different collision approaches. The experiment results leaned more to closed addressing than to open addressing and deemed linear probing impractical due to its low performance. Moreover, when items are randomly distributed with keys in a large space, different hash algorithms might produce similar performance. Increasing randomness in keys does not help hash table performance either and it seems that the load factor solely determines possibility of collision. These new discoveries might help programmers to design software products using hash tables.