Monte Carlo is a common method in financial engineering for a wide variety of problems, one being option pricing. Growing volumes and complexity of work that needs to be performed as well as strict requirements for fast responses makes for a pressing demand for high performance computing.

In this paper we study the applicability of GPUs for pricing exotic options (basket, Asian) using Monte Carlo techniques within the environment of a large financial institution. Contrary to many experimental results achieved by analyzing the computational kernel we focus on the performance of the end-to-end system that is now in production use at Handelsbanken, one of the major Swedish financial institutes.

We show that graphics cards can outperform CPUs given certain conditions and for reasonable problem sizes we find a 12x improvement over sequential code when pricing options in a production system.