HP updates Vertica for Big Data

Hewlett-Packard's Vertica subsidiary has updated its real-time analytics software, giving it a graphical user interface and connectivity to big-data-styled analysis systems.

The Vertica Analytics Platform is a column oriented analytic database, one designed for rapidly ingesting and structuring large amounts of data for quick analysis. "Our use case is focused primarily on real-time analytics," said Scott Howser, Vertica's vice president of product marketing. Internet companies such as Zynga and GroupOn both use Vertica for quick analysis of user behaviors, he said; he expects that an increasing number of organizations will require this type of immediate analysis to better serve their customers.

Integrating Vertica with Autonomy

HP purchased Vertica last year. It has since integrated Vertica with its Autonomy IDOL software, another 2011 HP acquisition, for a single solution, called the HP Next Generation Information Platform, that analyzes both structured and unstructured data.

Vertica 5.1, released Tuesday, features a new graphical user interface, as well as various improvements to its third-party drivers and access protocols, Howser said.

Deploying the new console, users can execute actions such as set up a Vertica cluster, add nodes, and drop databases. It also includes a dashboard for monitoring system performance. The console can be run from a browser, and actions are completed by dragging and dropping icons to the appropriate locations.

The previous interface was based on a comparatively primitive xTerm terminal emulator, Howser said.

Vertica 5.1 also comes with a number of significant updates to the protocols and drivers. Both the ODBC (Open Database Connectivity) and JDBC (Java Database Connectivity) drivers have been completely written, allowing C and Java programmers, respectively, to more easily connect their applications to the Vertica database.

Vertica can exchange data with Hadoop

The software also includes connectors for the Apache Hadoop data processing framework, which will allow Vertica deployments to move data to, or receive data from, a Hadoop-based big data analysis system.

The integration with Hadoop may be a key one, because such big data analysis is still what HP's combined Vertica/IDOL offering is missing, noted Ovum enterprise software analyst Tony Baer.

"Content stores are too heavyweight for the heterogeneous low-density data types of the so-called unstructured data world," Baer said, referring to data such as log files, machine data and transcripts from social network chats. Eventually, HP will either have to purchase a software company to further address this need, or, more likely, partner with a software provider specializing in this data analysis, such as Splunk or Hadoop vendors Hortonworks or Cloudera.