Financial Risk Management (FRM) Application on AWS using Microsoft HPC 2012 Sp3(High Performance Computing) ( 2016-2019)

By Kishore Kesanapally, VP of Data Analytics/Data Science/Business, Wells Fargo

The Financial Risk management application is an income forecasting, Asset-Liability Management & Market Risk Analysis application which enables clients to model the assets, liabilities, and off-balance-sheet instruments. It measures the Risk of a Balance Sheet under Any Economic Scenario. It creates reports that effectively and concisely deliver the management needs to make profitable financial decisions.

In the financial organization, I have worked for, the FRM applications’ quarterly/annual balance sheet jobs have been running over 200 on-prem data center parallel servers, and it took approximately a month to complete the jobs. It is due to a lot of financial calculations and it uses the Monte Carlo method. 

The management is decided to optimize the balance sheet job execution and save the analyst time by moving to the cloud to utilize the features like scalability, reduction in costs, Improved performance, secure and speedy disaster recovery, greener and time-saving on maintenance. We have evaluated various cloud providers and decided to go with AWS cloud and Microsoft HPC (high performance computing) as it meets our requirement.

Microsoft High Performance Computing (HPC) allows FRM Users to solve complex, compute-intensive problems. FRM applications require high network performance, fast storage, large amounts of memory and very high compute capabilities or all of these. AWS enables you to increase the speed of FRM runs and reduce time-to-results by running HPC in the cloud and scaling to more significant numbers of parallel tasks. AWS helps reduce costs by providing CPU servers on-demand (Spot Instances), optimized for specific applications, and without the need for large capital investments. We automated provisioning EC2 HPC Spot instances clusters on AWS, attach and detach DB’s, compute nodes for patching and restore databases from AWS S3 storage using PowerShell Scripting

The Financial Risk Management application (FRM) is deployed on AWS using cloud formation templates with auto scaling of hundreds of computing cores using Microsoft HPC (high performance clustering). HPC uses parallel computing features on Windows HPC Server 2012. The term HPC (High Performance Computing) is virtually synonymous with supercomputers and parallel computing. After implementing this project, the balance sheet job took only 24 hours to complete. By moving to the cloud, the ROI (Return on investment) of the project (over the next five years) is estimated to be $100M. It includes labor costs, data center costs and server maintenance.