How (not) to scale Amazon RDS

I'm currently working on a webbased service hosted on AWS. It's quite common for the application to see large spikes of users for a brief time. Because of this, our database has to withstand immense loads.

So far, our model was to use a single beefy RDS Postgres database instance, simply because we didn't need to care about the size of the database back when the project started. As time went by, the amount of visitors grew and grew. In order to keep up with the load, we just kept on throwing more money at the beast that is AWS by using larger and larger database instances. This had to stop. Everyone has heard of autoscaling database and compute instances before, so I dove right into the rabbithole of "the hyperscalable cloud" (tm).

First Attempt:

If you enjoyed this post, consider buying me a coffee! Got comments? Send me a Mail, or shoot me a message on Matrix.

Garrit Franke

Random thoughts, tips and rants about software


Website
Github
LinkedIn
Email
Matrix
PGP
RSS