Those guys from Seattle made a service called Fabric.
It does lots with data and we all think it's magic.
But there are some aspects where people get lost.
For example when asking "How much will it cost?"
Will your dev and production teams keep up their smile,
If you buy your compute in one great big pile.
Perhaps you will find there's a comfier fit,
if every department's service is split?
If I start off with one and then buy a new SKU,
Do I have one lake or do I have two?
Can you save lots of money get a round of applause?
If at the weekends your compute you do pause?
Do I need F2 or will I need more?
What is so special about F64?
Should we share it with others will performance be hit,
When someone writes code that's a little bit...
sub-optimal?
There's no one solution for a Fabric adoptions,
but once we are finished you'll know all the options.
When we've shared with our demos and shared with our voice,
you can all head for home set to make a good choice.