2 Minutes on: Decoupled Storage And Compute on Firebolt's Cloud Data Warehouse

Welcome to this short video on why decoupled storage and computing so great and how it works in Firebolt.

The nice thing about the coupled storage in compute is that when I will run the script I will be not able to just get an answer very quickly but actually choose the type of hardware i'm running this on.

Behind the scenes we don't just have databases but we have a concept called
engines. An engine in firebolt is the compute resource, the cluster ,you're using to run your queries.

When you define an engine or when you edit an
Engine you can easily choose the scale, this means how many nodes participate
in the cluster.

For more advanced users we can optimize further by literally choosing the ec2 instance type, so you can control things like how much rom and ssd your cluster has and your databases can have
multiple such engines attached to them

What about if i want to save a little bit of money and maybe run on a less expensive piece of hardware?

There are trade-offs you can make when you work on Firebolt. in such an environment you can decide for each query how many resources do you want to use and how much do you want to spend.

And so forth you never have to migrate to a stronger or cheaper cluster you don't have to re-cluster
re-index partition you don't have to worry about anything.

Just run your queries, choose your hardware with a few clicks and get going!