To transcend these limitations, Greenplum assembled a team of the world’s leading database experts and built a shared-nothing massively parallel processing database, designed from the ground up to achieve the highest levels of parallelism and efficiency for complex BI and analytical processing. In this architecture, each unit acts as a self-containe database management system that owns and manages a distinct portion of the overall data. The system automatically distributes data and parallelizes query workloads across all available hardware. The Greenplum Database shared-nothing architecture separates the physical storage of data into small units on individual segment servers (Figure 2), each with a.
