At this stage of its evolution, big data has been primarily used for batch MapReduce processing – massive amounts of data are collected for data scientists and business analysts to analyze for key insights. This works fine when looking at historical data or making far-forward predictions, but what about decisions that need to be made in real time? This is where in-memory solutions become important.
Data grid software leader ScaleOut Software is using in-memory computing to achieve operational analytics for real-time decision making. There are real benefits in support of in-memory technology when it comes to fraud alerts, transportation management, and taking advantage of short-lived financial opportunities.
ScaleOut’s in-memory data grid allows MapReduce to run 40x faster than Hadoop, giving users the ability to analyze data in real time. The company has a very compelling use case demonstration in the financial sector where a hedgefund uses operational analytics for some impressive results.
In the video below, watch a demo of ScaleOut hServer (in-memory execution engine for Hadoop MapReduce) & Apache Hadoop running a hedge fund strategies application. The video explains how this demo is an example of using ScaleOut hServer to run standard Hadoop MapReduce code over live, fast-changing data, a use case Hadoop traditionally cannot deliver. Moreover, ScaleOut hServer shows a 40X performance improvement over standard Hadoop MapReduce while running the demo.
Sign up for the free insideBIGDATA newsletter.