One day I’ll write a book „how to build scalable system for the 95%“.
You’ll be able to summarize it with
- pick any mainstream language
- deploy on one server
- deploy Postgres on another server
- double the number of server for redundancy
So many systems I’ve seen lately could run off in memory SQLite and a replication to a CSV file.
Yet there’s a relational database, a document database, a key value store and four services deployed on k8s for all 100MB data and 2000 customers.
@fallenhitokiri the industry is still just repeating variations of the premise that led to the classic post "Command-line Tools can be 235x faster than your Hadoop Cluster" <https://adamdrake.com/command-line-tools-can-be-235x-faster-than-your-hadoop-cluster.html>
Hadoop! Remember the hype? ~10 years hasn't resulted in the vanquish of hype-driven development. The industry is eternally fashion-first (and actual, real world requirements come second).