^
Advanced Search

Maxeler Technologies: Financial Computing System

Key Documents

Financial organizations can assess value and risk 30x faster on Xilinx-accelerated systems than those using standard multi-core processors.

As portfolios of complicated credit instruments have expanded, the process of valuing them and managing the risk has grown to a point where financial organizations use thousands of CPU cores to calculate value and risk daily. This task in turn requires vast amounts of electricity for both power and cooling. Not surprisingly, large financial institutions are searching for a way to add ever-greater computational power at a much lower operating cost.

To this end, the Applied Analytics group at J.P. Morgan in London launched a collaborative project with the acceleration-solutions provider Maxeler Technologies that remains ongoing. The core of the project was the design and delivery of a MaxRack hybrid (Xilinx® FPGA and Intel CPU) cluster solution by Maxeler.

The system has demonstrated more than 31x acceleration of the valuation for a substantial portfolio of complex credit derivatives compared with an identically sized system that uses only eight-core Intel CPUs. The project reduces operational expenditures more than 30-fold by building a customized high-efficiency high-performance computing system.

The Maxeler system clearly demonstrates the feasibility of using FPGA technology to significantly accelerate computations for the finance industry and, in particular, complex credit derivatives… One of the key insights of the project has been the substantial benefits to be gained from changing the computer to fit the algorithm, not changing the algorithm to fit the computer (which would be the usual approach using standard CPU cores). 

- James Spooner, Head of Acceleration (Finance), Maxeler Technologies

Additional Information

Quick Links