The company has Quantitative Analysts who run complex Monte Carlo simulations on deal pricing. The analysts evaluate multiple scenarios of how deals may evolve over the years based on market movements. This information is used by Traders to price, or take a market positions in the futures and options markets.
The simulations were being performed on high performance desktop PCs and took several hours to run. The company was looking at ways to improve accuracy and reliability of these simulations to enable traders to make more informed market decisions.
Avanade leveraged its knowledge of the financial services sector and cloud services capabilities to offer a more creative approach to the simulations. The Analysts submit the calculation from their desktop. Behind the scenes the calculation is split up in to parallel workloads, submitted to Windows Server High Performance Computing which manages data upload and jobs to Windows Azure. The Avanade team had pre-configured Windows Server HPC to manage on-demand provisioning.
By bursting workloads to the cloud, the company has reduced the time taken to process pricing simulation from more than 12 hours to less than 30 minutes. Automating the process that sends pricing data to Microsoft Azure provides a secure “real time” simulation and ensures that no sensitive data remains once the simulation is complete.