Minimal Data Stack

[dbeatty10]

What topic would you like to see discussed more in-depth?

I want to learn how I can set up a minimally viable collection of the modern data stack.

Because the "Arcane Acronym Game (AAG)" is so absurd that it is fun, here's some naming ideas:

• My Data Stack (MDS)
• Minimal Data Stack (MDS)
• Minimalist Data Stack (MDS)
• Minimal Viable Data Stack (MVDS)
• Major Doug Silliness (MDS)
• My Minimal Viable Data Stack (MMVDS)
• Essential Data Stack (EDS)
• Essentialist Data Stack (EDS)

Why do you think this is important?

As an analytics engineer, I know how to spin up dbt-core with an adapter to my favorite database and do some amazing things.

But it is unclear to me:

• Which portions of the modern datat stack (MDS) would give me the most value? Do I need all of them?
• Which other portions can I spin up in a permissionless fashion?
• How would I go about doing so?

Are there any existing resources that you’ve seen explore this topic? What do you think they’re missing? Please share any relevant links.

• Emerging Architectures for Modern Data Infrastructure describes a set of data infrastructure architectures.

What would happen beyond your wildest dreams?

I want to learn so much that I'm able to expand the state of the art and add some unique piece, yielding:

• More Modern Data Stack (MMDS)

Like the analytics version of cranking it up to 11.