Unstructured or 'Big Data' is data that comes in different shapes and sizes, is frequently changing, can be very short-lived, generally arrives in huge volumes, and originates from a broad variety of sources such as experimentation, web stream, point-of-sale, RFID tags, sensor arrays, etc. Unstructured data is very difficult to analyze and manage using traditional means such as relational databases.

SGI provides an end-to-end Big Data architecture, scale-up or scale-out, with the focus on both speed and scale. Created from best of breed components versus a proprietary stack, SGI Big Data solutions provide mission-critical, real time as well as batch capabilities. All solutions are factory-integrated and tested, providing a complete solution, and allowing you to start small and grow as your needs grow. The solution is X86- and Linux-based and can run any application and we have partners from data movement through analysis.

SGI has a heritage in this area, having deployed the largest Hadoop installations, and the largest single Hadoop cluster. We have SGI® UV™ 2000 solutions that gulp 64 TB databases into in-memory solutions that are orders of magnitude faster than flash solutions, capable of ingesting 4 TB/sec of incoming data. And when the data needs to be stored we have 10's of PB of storage safely stored under SGI® DMF™ software that can create active archives with our zero-watt ArcFiniti™ storage solutions.


SGI® DataRaptor™ with MarkLogic® Database SGI® Hadoop Solutions
  • Fastest way to deploy and scale MarkLogic operational database
  • Plug In and GO
  • Pre-configured, optimized, and integrated Starter Kits
  • Custom Design-to-Order Clusters