Splunk Delivers Hunk: Analytics for Hadoop

By Darryl K. Taft  |  Posted 2013-10-29

Splunk Delivers Hunk: Analytics for Hadoop

Splunk, provider of a popular software platform for real-time operational intelligence, has announced the general availability of Hunk: Splunk Analytics for Hadoop.

Initially released in beta at the Hadoop Summit North America 2013 conference in San Jose, Calif., in June, Hunk is a full-featured, integrated analytics platform for Hadoop that enables users to interactively explore, analyze and visualize historical data in Hadoop.

Essentially, Hunk is a software product from Splunk that integrates exploration, analysis and visualization of data in Hadoop. It enables users to interact with and analyze data in Hadoop without programming, integrations or forced data migrations, the company said.

Indeed, Splunk officials said Hunk is built on patent-pending virtual index technology that delivers self-serve analytics without the need for specialized programming skill sets, fixed schemas or costly integrations.

Hunk also provides a rich developer environment, including an integrated Web framework and software development kits (SDKs) for the most popular development languages, including Java, Python, Ruby, C#, JavaScript and PHP.

"With Hunk, it's not about re-tooling or learning new things; it's about using the Web skills you have," said Jon Rooney, director of developer marketing at Splunk. "It's really about enabling developers where they are," he added.

“Hunk is transforming the way organizations analyze their data in Hadoop by replacing drawn out development cycles with software that enables customers to deploy and deliver insights in hours instead of weeks or months,” said Sanjay Mehta, vice president of product marketing at Splunk, in a statement.

Hunk is available for download for a free 60-day trial with no caps on data size or number of Hadoop nodes.

“Hadoop is an increasingly important technology and many organizations are storing vast amounts of data in Hadoop," Mehta said. "However, this often creates a problem because the data sets become too big to move and more traditional approaches to analytics of raw data in Hadoop require brittle, fixed schemas. These are key reasons our customers consistently tell us about the cost, time and sheer difficulty of getting analytics out of their Hadoop clusters. With Hunk, we applied everything Splunk has learned from 10 years of experience with more than 6,000 customers to this unique challenge.”

Splunk announced the general availability of Hunk on Oct. 29 jointly at its SplunkLive! event in San Francisco and at the O'Reilly Strata Conference + Hadoop World 2013 show in New York City.

“Splunk was early in understanding the importance of machine-generated data, and in building an analytic stack focused on it, with strengths in quasi-real-time interfaces and in schema-on-need handling of time series data,” said Curt Monash, president of Monash Research and editor of DBMS 2, in a statement. “Now Splunk’s analytic tools are also available for machine-generated data stored in Hadoop. Splunk has further strengthened its schema-on-need story by adding its new higher-performance data store.”

Splunk Delivers Hunk: Analytics for Hadoop

Mehta said more than 100 organizations took part in the Hunk beta program. Beta participants consistently reported simple installation and fast analytics results on data in Hadoop.

“Hunk lets us quickly and easily visualize data in our Hadoop clusters,” said Hideaki Sakaue, executive officer at Hitachi Solutions, in a statement. “Hunk removes the need for specialized skill sets and lengthy development cycles so organizations can focus on innovation. We look forward to delivering this value to our customers.”

Hunk enables users to quickly derive actionable insights from raw, unstructured and polystructured data in Hadoop, said Clint Sharp, director of product management, big data and operational intelligence at Splunk. Hunk works on Apache Hadoop and most major distributions, including those from Cloudera, Hortonworks, IBM, MapR and Pivotal.

Product features include integrated analytics, fast deployment and an interactive search and results preview. The preview capability enables users to interact with data, change perspectives on-the-fly and preview results as MapReduce jobs are running. Hunk’s schema on-the-fly delivers the flexibility to run queries on Hadoop data without the need to know up front what questions or reports will need to be asked or produced later.

“Hunk enables our enterprise customers to achieve their big data goals,” said Kou Miyake, president and CEO of NTT DATA INTELLILINK, in a statement. “Hunk accelerates insights from Hadoop with a much faster time-to-value than open-source alternatives. Hunk also enables enterprise developers to build big data applications because of the rich developer environment and tooling.”

Hunk also offers drag-and-drop analytics with the patent-pending Splunk Data Models and Pivot Interface, first released in Splunk Enterprise 6.

“Hunk gives business analytics teams using Hadoop in their stack an enormous opportunity to improve overall efficiency for everyone–the analysts can more quickly explore data and create visualizations, the operations teams can better monitor data security and developers can better ensure data quality,” said Marcus Buda, senior data architect at the Otto Group, in a statement. “If you are working in an agile collaborative way, there are tremendous synergies in monitoring your data processing pipelines for Hadoop and at the same time using your knowledge about that data to develop business analytics dashboards with Hunk. The combination of Splunk software with the processing power of MapReduce and Hadoop generates a transparent and sustainable competitive advantage for your company within minutes.”

Pricing for a one-year term license of Hunk starts at $2,500 per Hadoop node with a minimum of 10 nodes.

Rocket Fuel