Falcon Acceleration for Finance Applications
Acceleration of Financial Workloads by up to 35X
In the field of finance, high performance computing is useful solving many problems. Financial applications demand efficient algorithms along with high-speed computing to solve real-time workloads. Applications such as high-frequency trading, complex simulations and real-time analytics are essential to the success of capital markets brokerages and other financial firms.
After the emergence of Black-Scholes, the first widely used model for option pricing, complex mathematic models, numerical processing, and large-scale computing in computational finance has become more universal. However, compute performance, such as memory capacity and CPU speed has become a limiting factor for real-time applications such as:
Parallelized algorithms on FPGA acceleration platforms, such as Monte Carlo simulation, can enable financial institutions to shift from overnight to intraday risk calculations, reducing overall exposure to the market.
Use existing C/C++
No need to rewrite working algorithms, just recompile with the Merlin Complier for your choice of FPGA
Faster results with Lower Opex
up to 25X better performance per watt and 50-75x latency improvement compared to CPU/GPU implementations.
Fortunately, with the rapid development of heterogenous public and private data centers, featuring both
CPU andFPGA acceleration, algorithms such as Black Scholes can be accelerated up to 35X leveraging
FPGAs. Falcon Computing’s Merlin Complier makes porting C/C++ algorithms from a CPU to an FPGA
a snap. With no need for rewrites, you get all the performance in a fraction of the effort.
Merlin Speed ups over CPU
Today with Merlin Compiler
in the Cloud.