Hadoop growing, not replacing RDBMS in enterprises

In most cases, Hadoop coexisting with conventional relational database management tools, a new study says

The growing need for companies to manage surging volumes of structured and unstructured data is continuing to propel enterprise use of open-source Apache Hadoop software.

However, instead of replacing existing technologies, Hadoop appears to be finding more of a place working alongside conventional relational database management systems, according to a new report from Ventana Research, a business and IT advisory firm in San Ramon, Calif.

Hadoop is designed to help companies manage and process petabytes of data. Much of the technology's appeal lies in its ability to break up very large data sets into smaller data blocks that are then distributed across a cluster of commodity hardware for faster processing.

Early adopters of the technology, including Facebook, Amazon, eBay and Yahoo, have been using Hadoop to store and analyze petabytes of unstructured data that conventional RDBMS setups couldn't handle easily.

Ventana's report, which is based on a survey of more than 160 companies, shows that a growing number of enterprises have begun putting Hadoop to use for similar purposes.

More than half of all enterprises looking to glean business insights by analyzing very large volumes of structured and unstructured data are said to have begun using Hadoop to help them with the task.

Most are using Hadoop to add new capabilities rather than to replace existing technologies, said David Menninger, author of the Ventana report.

Ventana's research shows that a majority of companies that are using Hadoop are using it mainly to collect and analyze huge volumes of unstructured and machine-generated information, such as log and event data, search engine results, and text and multimedia content from social media sites.

The technology is much less likely to be used for analyzing conventional structured data such as transaction data, customer information and call records, where traditional RDBMS tools still appear to have an edge, Menninger said.

People are using Hadoop because it enables new kinds of data analytics capabilities, he said. "In two-thirds of the cases, we found that people are using Hadoop for advanced analytics, and for types of analysis that they were not doing before."

Hadoop appears to be delivering new capabilities, especially on the operations side of the business, Menninger said. In most cases, Hadoop is being used by business units such as sales and marketing rather than by groups like human resources and finance, he said.

"Operations is the place where the most up-to-the-minute and granular data occurs. It is a place where a lot of the data is machine-generated," he said. "It is also the group which is asked to support other areas of the business."

Despite Hadoop's early promise, enterprises still face some significant challenges in adopting the open-source tool. One of the biggest problems continues to be the relative shortage of people who are skilled with Hadoop, Menninger said.

The obstacles that were cited by survey respondents most often related to staff availability and training, Menninger said. A lot of companies also appear to be having problems dealing with Hadoop's clustered computing approach.

Enterprises also appear to be a bit unsure about security when it comes to Hadoop. Just 49% of the respondents to Ventana's survey said they were satisfied with Hadoop's security.

Jaikumar Vijayan covers data security and privacy issues, financial services security and e-voting for Computerworld. Follow Jaikumar on Twitter at  @jaivijayan, or subscribe to Jaikumar's RSS feed . His email address is jvijayan@computerworld.com.

Copyright © 2011 IDG Communications, Inc.

7 inconvenient truths about the hybrid work trend
Shop Tech Products at Amazon