Answers are grounded in this paper and cite their sources.
Greenplum Database (GPDB) has many features designed to enable data scientists. Before a data scientist can use GPDB, a database administrator (DBA) must provision a cluster and install any required data science packages. Provisioning a GPDB cluster on bare metal requires a lengthy setup process. Scaling, recovering, and securing the cluster post-deployment are also complex. Greenplum for Kubernetes (GP4K) abstracts away these complexities, simplifying and automating the process for users. In this demonstration, we introduce GP4K with an opinionated deployment and a declarative manifest. We provide a brief overview of GP4K's architecture and discuss its implementation. We also demonstrate a full life cycle of managing a cluster from birth to retirement, including scale-up and self-healing all with minimal DBA inputs.
No takes yet. Share an insight, caveat, or question.
patel et al. (2019) studied this question.