Scale As it Makes Sense for Your Business
F3 connects with standard market data sources including Bloomberg and Thomson Reuters and your applications connect through APIs using Java, C++, C#, Python and other languages, to achieve results fast. Additionally, F3 enables parallel calculations in any combination of on-site servers and the Cloud. Calculation results are stored in memory and are only recalculated when underyling data or trades change, ensuring resources are managed efficiently.
F3 is a highly scalable analytics solution that can be deployed in a dedicated environment optimized specifically for risk analytics or as a technology platform for managing data consumed and created by F3’s analytics. Whether deployed on the desktop, across departments, or across the enterprise, F3 scales with your organization as your needs evolve. Additional processing power can be added on demand without having to modify the configuration of F3 Platform; and processing nodes can be simultaneously deployed both locally and in the cloud.
F3 Platform is built with reliability in mind. It is designed to run on Windows Server® and Linux® operating systems with support for enterprise-grade relational databases and web servers.
High availability is built into the heart of F3 Platform. All F3 services can be distributed across multiple machines. In the event of a calculation server failure, F3 Platform reroutes calculations to other available servers. F3 Platform works with web servers that support load balancing and hot failover. F3 Platform works with database clustering on all supported database products.
F3 centrally manages simulations and valuations on large portfolios by running calculations in parallel, enabling intraday analysis of large, multi-asset, multi-currency portfolios. With F3 Platform, computationally intensive simulations for calculations such as CVA and VaR can be distributed across grids of local machines and both private and public cloud environments. F3 Platform manages the complexities of decomposing large calculations, distributing them across multiple processors, recombining the results, and delivering them in a consistent and coherent manner to the calling system.