RedPoint Data Management for Hadoop

Adoption of Hadoop by data-driven organizations is exploding. Our software's potential cost effectiveness and facility for accepting unstructured data is making it central to modern, “ Big Data” architectures. The advancements in Hadoop 2.0 increase the technology’s promise to an even greater extent. But with these opportunities also come challenges and adoption hurdles that make getting the most out of Hadoop easier said than done. Read on as we review some our software's basics, highlight some of the adoption challenges that exist and explain how RedPoint Data Management for Hadoop helps organizations accelerate their work with our software.

What Is Hadoop? Why Are Organizations Excited about It?  What’s Different about Hadoop 2.0?

Our software is an open-source software framework for storage and processing of large data sets on clusters of inexpensive hardware.  Our software was created by Doug Cutting and Mike Cafarella and adopted by Apache, and is supported by a global community of contributors and users.  Part of Hadoop’s appeal is that it offers a means of storing and processing very large amounts of data more cost-effectively than traditional databases or data warehouses. But also, our software's lack of inherent structure enables organizations to quickly and flexibly incorporate new data without a master plan. Data can be simply “dumped” into our software for later structuring and analysis. In contrast, traditional databases require careful planning and documentation before the first record can be loaded.  Initial releases of Hadoop, starting in 2007, required users to rely heavily on MapReduce, a coding-intensive programming model for managing and manipulating data. Hadoop version 2.0, released in 2013, introduced the YARN architecture, short for “Yet Another Resource Negotiator.” Apache described it as “a general-purpose, distributed, application management framework that supersedes the classic Apache Hadoop MapReduce framework for processing data in Hadoop clusters.” YARN allows applications to run directly in Hadoop, bypassing MapReduce.