- NVIDIA GRID
- 3D Vision
- About NVIDIA
The financial derivatives market is a high-stakes game, where even the slightest error in the valuation of a contract can lead to
big losses. Consequently, traders rely on complex mathematical models to arrive at the value and risk sensitivity of the contract.
The quick-paced nature of the financial markets makes it imperative that derivative valuations be fast and accurate.
SciFinance is a derivative model development environment that automatically generates serial C/C++ source code from
concise, high-level model specifications. Now, in order to dramatically speed up the execution time of the Monte Carlo
pricing and risk models, SciComp has added a feature that automatically generates NVIDIA® CUDA™-enabled
source code. This new code style allows critical sections of the pricing code to take advantage of the highly parallel
architecture of NVIDIA GPUs. To trigger the new code style, customers need simply add the keyword ‘CUDA’ to a model specification to produce
CUDA-enabled, compiler-ready parallel code. The result: speed increases from 30X to over 100X with one NVIDIA Tesla C1060 GPU. Further increases are nearly linear with the number of GPUs.
The ability to create and execute Monte Carlo pricing models much faster allows traders and risk managers to assess alternative modeling scenarios and enhance risk analysis. The improved understanding of the derivative contract and its risk sensitivities increases the profit potential of the deal.