Maximum clique solver using bitsets on GPUs

Abstract

Finding the maximum clique in a graph is useful for solving problems in many real world applications. However the problem is classified as NP-hard, thus making it very difficult to solve for large and dense graphs. This paper presents one of the only exact maximum clique solvers that takes advantage of the parallelism of Graphical Processing Units (GPUs). The algorithm makes use of bitsets to reduce the amount of storage space needed and take advantage of bit-level parallelism in hardware to increase performance. The results show that the GPU implementation of the algorithm performs better than the corresponding sequential algorithm in almost all cases; performance gains tend to be more prominent on larger graph sizes that can be solved using more levels of parallelism.