Accelerate Financial Engineering Models

Financial Engineering

Which approach provides the most applicable acceleration in Financial Engineering? Porting to GPU, use of the Techila Distributed Computing Engine (TDCE) and power from Windows Azure cloud platform, or a native rewrite for a FPGA? Christos Delivorias answers these questions in a University of Manchester Seminar on implementing financial models on GPUs, FPGAs and in the Cloud.

The case studies in acceleration of the Heston Stochastic Volatility Financial Engineering Model show that the closer to the machine level the implementation is the better the acceleration achieved. On the other hand, plain speed up factor does not provide the full picture. TCO is a key metric for many businesses and researchers.

The handicap with the specialized development approaches are the high friction and niche specialization in the implementation of code. The cost of acceleration is an important factor, and should be taken into account when assessing different solutions. This is where the Techila Distributed Computing Engine is superior compared to the specialized development approaches. If your code is under constant development, ease of development will be a great benefit.

An interesting observation in the research presented in the University of Manchester is that a FPGA becomes saturated and plateau at a certain acceleration. Techila Distributed Computing Engine and GPU are still increasing.

More news