Financial services institutions are increasingly relying on real time analytics for fraud detection, digital identity management and many other areas. Irrespective of the technology underlying the analytics, one fundamental premise remains true: without having sufficient data to power these algorithms and the ability to load that data in real time, even the best algorithms will return suboptimal results. Having worked on multiple custom fraud solutions, Aerospike's Chief Dev Advocate Tim Faulkes will explore these data challenges in performing fraud detection, then look at models and technology which can be used to address this. Whilst the modeling techniques used will focus primarily on credit card fraud, similar techniques can be applied to other areas requiring real-time access to large quantities of data.