ScaleOut Software has revamped its in-memory data grid platform to deliver a twentyfold increase in MapReduce performance while eliminating some of the costs associated with traditional Hadoop clusters. The newly launched hServer V2 is specifically geared toward ad hoc analysis of operational data.

While not designed to compete with offerings from distributors such as Cloudera and Hortonworks, the latest release of hServer utilizes select Hadoop components to greatly reduce deployment times and provide native support for “most” Java-based MapReduce applications. The solution can be implemented with ScaleOut’s integrated in-memory data grid or external data repositories such as HDFS.

The company claims that the solution can be used to simplify the development of Big Data apps and accelerate analytics across wide range of industries, including financial services and e-commerce.

Bill Bain, the chief executive officer of ScaleOut, said that “this release marks a huge step forward for real-time data analytics. By enabling real-time analytics for Hadoop, which has emerged as by far the most popular platform for analyzing big data, we aim to dramatically improve the effectiveness with which organizations can manage their live data. Reducing Hadoop’s execution time by more than an order of magnitude will make a tremendous difference in the ability to better understand – and predict – key patterns and trends with live, fast-changing data.”

Last month at Oracle OpenWorld 2013, Larry Ellison unveiled the In-Memory Option Database 12c. The CEO said that the function speeds up querying by 100 times to facilitate the rapid processing of transactional and analytical workloads.

Hailed as fast and simple by Oracle, the In-Memory Option has come under criticism for being impractical from a cost standpoint.