What is unique about Greenplum Database architecture?

posted Oct 3, 2012, 6:04 AM by Sachchida Ojha   [ updated Oct 3, 2012, 6:04 AM ]
Greenplum Database is a shared-nothing MPP (massively parallel processing) architecture designed from the ground up for petabyte-scale analytical workloads and ad-hoc analytical queries. In this architecture, data is automatically partitioned across multiple ‘segment’ servers, and each ‘segment’ owns and manages a distinct portion of the overall data. All communication is via a network interconnect– there is no disk-level sharing or contention to be concerned with (i.e. it is a ‘shared-nothing’ architecture).