Achieve stunning acceleration for PDE and Monte Carlo derivative pricing models

SciFinance®, the ultimate in flexible derivatives pricing model development, has been enhanced with the SciGPU component for generating GPU-enabled or OpenMP source code.

SciFinance automatically generates NVIDIA CUDA-enabled source code for derivatives pricing models. No CUDA or parallel programming expertise is required.

Typical speedups:

• CUDA Monte Carlo pricing models:
  • 30X-50X faster than serial code (single GPU, double precision)
• CUDA PDE pricing models
  • 10X-35X faster than serial code (single GPU, double precision)
• OpenMP-compliant code executes in the multi-processor environment with nearly linear speed-up
  • 3.9X faster than serial code on a quad-core PC, 22X on a 24 CPU workstation

>Download the SciComp / NVIDIA Tech Brief to learn more How does SciFinance work? Select from hundreds of provided pricing model specifications Modify the model specification as needed Test the serial version of the pricing model Add keyword "CUDA" to the specification and re-synthesize SciFinance does the rest by automatically generating fully documented GPU-enabled pricing model source code

Why use SciFinance for GPU computing?

• SciFinance automatically generates NVIDIA GPU-enabled or OpenMP-compliant source code for any financial derivative that can be valued using any system of partial differential equations (PDEs) or stochastic differential equation (SDEs, i.e., Monte Carlo)
• No CUDA or parallel programming expertise required
• SciFinance automatically generates optimized CUDA code
• Optimized for double precision
• Mixed single/double precision code style option available
• Supports complex models with large parameter lists while respecting the CUDA kernel arguments limit
• Efficient CUDA local memory management
• Multi-GPU support

> Brief Overview Movie of GPU Computing with SciFinance

> SciFinance-generated Derivatives Pricing Model Examples

 

Case Studies

>Download the Reval/SciComp Case Study In order to quickly develop new structured products and vastly speed up Monte Carlo-based derivatives for its Software-as-a-Service (SaaS) customers, Reval uses SciFinance.