This blog is about the surprise costs of using Google Cloud’s services. I say ‘surprise’, however I failed to research Google’s cloud costs ahead of time, so it’s really my fault. 🙂
Our capstone project team decided to use Google Cloud PostgreSQL for our database. We built the data outline and model, stood up the virtual instance, and loaded the dump file successfully (after a few tries). Everything seemed good and well, so I let the database instance sit idly with 0 activity.
After only 2 weeks later, I logged back into GCP and noticed that my billing account had an issue: our billing credits were depleted. Upon further investigation I found that our PostgreSQL database instance incurred over $30, and was forecast to exceed $70 for the month! This was quite shocking, because my database activity was limited to:
- Standing up the database
- Loading the dump file
- Letting the database sit idly
I did some research online (Reddit, Medium, Stack Overflow) and found that other people had similar experiences. It turns out, the database configurations that I chose (the default settings) were much more than I needed for the capstone project.
I then found another article that provided instructions on how to setup a micro instance database with an estimated cost of $9.37 per month (I’ll believe it when I see it). I then deleted the old database and then stood up a new database with the micro instance configuration settings.
So, this was an expensive (not really) lesson learned in cloud services in costs. I’m sure this is a lesson that I can carry forward throughout my career: know what you’re paying for. :p
