LexisNexis Risk Solutions, formerly ThreatMetrix, gives customers the ability to support better end customer experience while keeping fraud volumes low. Each day, when a customer requests a risk decision from the decision engine, it uses information crowdsourced from 5000 companies globally to:

• Handles over 130 million real time transactions across 40,000 websites to quickly determine which requests are legitimate customers versus cyber threats.

• Processes two billion API transactions per month, from 4.5 billion active devices.

• Performs calculations that include anonymized device profiling, smart rules, smart learning, and data from hundreds of properties stored in the data warehouse.

• Returns hundreds of attributes that inform the customer whether the request is a cyberthreat or a valued customer in less than 100 milliseconds.

As global digital transformations accelerate and the volume of transaction requests, connected devices, and online activity grows, LexisNexis Risk Solutions’ transaction capabilities and performance must stay predictable. Their original solution - Cassandra - required too many nodes and too much overhead to gain desired performance, especially as traffic and data volumes scaled up.

With Aerospike, LexisNexis Risk Solutions eliminates the risk that they experienced with Cassandra of unknown costs associated with uncontrolled scale. Aerospike provides absolutely predictable sub-millisecond latency performance despite dramatically increased volumes of request, throughput requirements, and a limited operations team. With Aerospike’s hybrid memory architecture that keeps the index in-memory and data on SSD, LexisNexis Risk Solutions can handle hundreds of thousands of reads and writes per second, replication across multiple datacenters with minimal latency, and provide customer decisions globally to help distinguish real requests from cyber threats.

When LexisNexis Risk Solutions hit performance and scaling challenges with legacy solution Cassandra, they moved to Aerospike for a solution that was easier to manage, saved costs, and delivered predictable performance even as their data needs scaled.

For a deeper understanding and more insights, explore these additional resources.