AWS launches PostgreSQL Serverless Service

Amazon just announced Amazon Aurora PostgreSQL Serverless service

The database is usually the most critical part of software architecture and managing databases, especially relational ones, has never been easy. For this reason, we created Amazon Aurora Serverless, an auto-scaling version of Amazon Aurora that automatically starts up, shuts down and scales up or down based on your application workload.

You won't need fixed RDS instances or EC2 instances with database servers attached to auto-scaling groups. This would be a good improvement regarding saving cost and scaling capabilities.

So, if you have a dev/qa environment you don't need to have a database running all the time, now using serverless service, it would be charged on-demand saving a lot of money.

For production purpose, you don't need to handle auto or manual scaling by yourself anymore. Handling traffic peaks would be as easier as any serverless(on-demand) service.

You can find everything related here

Comments

[ahmed sobhy elgawesh]

thank you for your post
and your simple and clear way

i didn't have a wide knowledge about AWS

but i want to know more about this sentences:

automatically starts up, shuts down and scales up or down based on your application workload.

it wasn't clear for me

from elastic description
-> What Is Elastic Load Balancing?
Elastic Load Balancing distributes incoming application or network traffic across multiple targets, such as Amazon EC2 instances, containers, and IP addresses, in multiple Availability Zones. Elastic Load Balancing scales your load balancer as traffic to your application changes over time, and can scale to the vast majority of workloads automatically.

so what are the advantages
and if the point that is automatically starts up, shuts down
what is the benefit in this case

thanks for your simple and clear post Levi
have a great day

[Levi Velázquez]

Hi,

Even using Elastic Load Balancing you would need to have at least one instance running no matter what, so, if you have 7 days without touching your qa environment, u will pay at least 1 instance for every day. Using serverless, you pay for what you use, so, 7 days without any database request is equal to 0$.

Besides that, you need to set up the rules for scaling up, like CPU usage, RAM usage, etc. That is not so trivial. Therefore, always you need to establish a maximum number of instances that could be running at a certain time according to the metrics. Maybe you will be short if it receives high traffic, with serverless you don't need to worry about that, everything scales automatically on-demand.

[ahmed sobhy elgawesh]

yes i got it
thanks for you so much