Distributed data(base), some simple experiments.
TL;DR: in distributed databases (Example: Yugabyte), it helps to know how to define your tables. The default behaviour is Optimised (sharded, distributed) for Very Large Tables. But Small tables also need attention. Too Much of a Good Thing... Background Distributed Databases are The Future. That is why I began to experiment with Yugabyte. I managed to create a 6-node (yes, Six Nodes) cluster in no time . And because Yugabyte is fully Postgres Compatible my good-old pg-scripts work straight away. From the install-story, I found that by default all my tables seem sharded over 6 tablets and that was something I wanted to investigate futher. So, Let's Play.... The Demo I needed some demo-tables first. With YB comes the "northwind" demo (link) . This demo was promptly installed from the command-prompt from any of the nodes (regardless of which node: in my case they are all equal). I shell-ed into the container, and typed # Yugabyted demo connect And there it was. I also us...