Research Library

Technical Insight: Persistent Storage for Apache Spark in the Enterprise

Published June 24th, 2016. Evaluating MapR’s Converged Data Platform in the context of Spark-based applications, by John Webster, Senior Analyst with Evaluator Group.

Evaluating MapR’s Converged Data Platform in the context of Spark-based applications

by John Webster

Apache Spark

Apache Spark follows a tradition established by Hadoop. It delivers data analytics applications using a distributed computing cluster designed for performance at large scale while costing significantly less than the traditional enterprise data warehouse. Spark is both fast and general purpose owing to its implementation of a more efficient code base vs Hadoop, and an in-memory processing architecture that accelerates performance while still leveraging commodity hardware and open source code. It is also easily accessible, supporting a broad range of APIs.

Download now to read the full report. 


Register or login to download this Free document

Forgot your password? Reset it here.