Install Greenplum Database on GCP!
Today i decided to try installing Greenplum Database on GCP via the marketplace offering from Pivotal on the Google Cloud Marketplace. The whole process was seamless and took about 10 minutes to build a cluster with 1 master and 4 worker nodes.
From the bookkeeping side, i had to create my GCP project to handle billing. After that was done, i navigating to the console and searched for marketplace and choose greenplum in the search. After selecting it a window came up to pick my options as shown below:
In this window of options i chose 1 master and 4 segments, there are many cluster size options to pick from. I chose HDD and not SDD, because this was really for a demo purpose and I dont need the speed. And then I selected to pre-install Greenplum Command Center but not the analytical modules.
After pressing go a window came up which took about 5 minutes to deploy the cluster. You can click on the cluster name and it will expand a tree view and show you the details of the installation process so you can track it. When this is done on the right side it has URLS, Hosts and Ports you need to connect to the Greenplum Database as shown below.
On the right you can see the necessary links and instructions to connect via SSH, gcloud, pgadmin or GPCC, including relevant credentials. I really enjoyed connecting to the cluster with gcloud. After installing gcloud on my system, i was able to link my google account with the Google CLI and then run a command like this from the terminal to get full SSH access to the cluster:
$ gcloud compute ssh gpadmin@pivotal-greenplum-novick-1-mdw
Awesome! Next step was to connect via the command center, just click on the link in the control panel and then add the credentials provided and command center opened right up as shown below.
Finally i wanted to pause my instance so i don’t get billed too much so i went into the Compute Engine menu and selected instances. From there i checked all the nodes that are part of the cluster and hit the stop button to pause the instances. I will come back to this cluster later. Check out the screen shot below.
This is a great way to provision Greenplum systems and I am going to use this method many times again in the future! Try it for yourself and leave a comment.
Working on enterprise software since 2002, and on big data and database management systems since 2007. Started on Greenplum Database in 2009 as a performance engineer and worked in various R&D and support capacities until shifting into product management for the world’s greatest database: Greenplum.