High Performance and Low Power Monte Carlo Methods to Option Pricing Models via High Level Design and Synthesis

Abstract

This article compares the performance and energy consumption of GPUs and FPGAs via implementing financial market models. The case studies used in this comparison are the Black-Scholes model and the Heston model for option pricing problems, which are analyzed numerically by Monte Carlo method. The algorithms are computationally intensive but not memory-intensive and thus well suited for FPGA implementation. High-level synthesis was performed starting from parallel models written in OpenCL and then various micro-architectures were explored and optimized on FPGAs. The final implementations of both models to several options on FPGAs achieved the best parallel acceleration systems, in terms of both performance-per-operation and energy-per-operation, compared not only to the kernels on advanced GPUs but also to the RTL implementations found in the literatures.