Over a million developers have joined DZone.

Greenplum Revs Internet-Scale Database

DZone's Guide to

Greenplum Revs Internet-Scale Database

· ·
Free Resource
Greenplum, the other (Postgres/Bizgres-based) open source database near and dear to Sun besides MySQL, this one capable of messaging a petabyte of data and buoyed by a recent $27 million infusion, has moved on to rev 3.0, a k a G3, claiming it can analyze vast amounts of business data 10 to 100 times faster than traditional data warehouse widgetry at a fraction of the cost.

It positions itself as the world’s first “Internet-scale” database capable of dealing with the Web 2.0-sized information overload because of its parallel processing, shared nothing architecture.

The 3.0 iteration is supposed to do petabyte loading (3.5TB/hour); natively support advanced parallel analytics; handle external data streams like real-time stock market data and RSS feeds, web pages and Web Services; manage large and diverse user groups; provide full SQL92 functionality; and interoperate with Business Objects, Cognos, IBM Data Stage, Informatica, MicroStrategy, Pentaho and SAS.

It claims to be reinventing the database for business intelligence.

Besides Sun and its data warehouse appliance, the five-year-old Greenplum counts Skype, Comcast, and VideoEgg among its customers; has Dell, IBM, HP, EMC and NetApp as vendor and folks like Capgemini and Bull as service providers.

It is targeting the $14 billion BI/data warehousing market and runs on any commodity x86 hardware, leveraging other open standards such as gigabit Ethernet and SATA disk drives.

Greenplum competes against Netezza, IBM, Oracle, Teradata and HP’s newfangled Neoview. It offers subscription-based per-terabyte pricing and discounts to start-ups.

Greenplum’s roadmap calls for the addition of a distributed query dispatcher, point-in-time recovery, disaster recovery enhancements, automated recovery to spare host/segment/pool and more in-engine analytics.

It says it wants to open up the engine to ISVs via APIs to improve scalability and performance by orders of magnitude.


Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}