SGI & Hadoop

SGI ANNOUNCES WORLD’S FIRST POWER OPTIMIZED HADOOP CLUSTER REFERENCE IMPLEMENTATION

Green Cloudera-Based Hadoop Cluster to Include Pre-Installed Business Intelligence Software from Key Industry Vendors

Fremont, Calif.— November 9, 2011 — SGI (NASDAQ: SGI), the trusted leader in technical computing, today announced its first SGI Hadoop Cluster Reference Implementation (RI) and the worlds' first power optimized Hadoop cluster during the Hadoop World Conference being held in New York City. The first product of its recently announced relationship with Cloudera, Inc., the SGI Hadoop Cluster RI allows customers to address the need for high performance data analytics while managing their power consumption, one of the key costs in total data center cost of ownership.

As customers analyze ever larger amounts of unstructured data to derive more value and ROI from their data streams, their Hadoop clusters scale and consume ever larger amounts of energy. The SGI Hadoop Cluster RI runs SGI Management Center software on servers utilizing Intel® Xeon® processors with Intelligent Power Node Manager (IPNM) to deliver cost-effective, full system power management. This technology combination enables customers to tune system power consumption to changing data center power envelopes and to tune system performance (operations/watt) to an optimal point for any given analytics workload.

“SGI has delivered tens of thousands of ‘design to order’ Hadoop cluster nodes optimized to meet our customer needs,” said Praveen Mandal, senior vice president of engineering at SGI. “The SGI Hadoop Cluster RI solution will allow customers who don't want to design an implementation from scratch or need the shortest time to production to easily acquire a complete factory integrated, ready to run Hadoop cluster.”

Based on SGI's expanding relationships with key business intelligence software vendors, SGI's Hadoop Reference Implementation also delivers a tested, optimized, ready-to-run Hadoop system with an ecosystem of analytical solutions which allow developers to more easily construct best-in-class, differentiated BI solutions. SGI is working with Kitenga Inc., Datameer, Inc., Pentaho Corporation and Quantum4D, Inc., to deliver the following differentiated analytical capabilities to customers in Federal, financial, social media, telecommunications, and other key markets:

• Quantum4D: data modeling and interactive data visualization capabilities on Hadoop for Actionable Insight;
• Kitenga: a new generation of Big Data insight engine with integrated search, information modeling and visualization capabilities;
• Datameer: allows for MS Excel-based analytical capabilities with the SGI Hadoop Reference Implementation and enables business users to access, analyze and visualize massive amounts data on dashboards;
• Pentaho: provides Pentaho Business Analytics on the SGI Hadoop RI, leveraging a graphical ETL environment for creating and managing Hadoop MapReduce jobs. This easily integrates data from other sources and provides end-to-end business analytics on Hadoop, including reporting, ad hoc query, interactive analysis and data integration.

Availability

The SGI Hadoop Cluster Reference Implementation is available for order now and is the first of several reference implementations that will be available in the coming months, addressing optimizations such as high performance and scale, and optimized analytics applications.

About SGI
SGI, the trusted leader in technical computing, is focused on helping customers solve their most demanding business and technology challenges. Visit sgi.com for more information.

Connect with SGI on Twitter (@sgi_corp), YouTube (youtube.com/sgicorp), and LinkedIn.

Contact Information:

Ogilvy Public Relations
Meghan Fintland

415-677-2704
SGImedia [at] ogilvyr [dot] com

© 2011 Silicon Graphics International Corp. All rights reserved. SGI is a registered trademarks or trademarks of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries. All other trademarks are property of their respective holders.