Deployment Pipelines with Embedded SKU

[sapientcompany]

Does anyone know if deployment pipelines share resources with an embeded SKU.  For example, if I have A3 and I setup a deployment pipeline, will each workspace created (Dev, Test etc...) share the A3 resource, meaning my production workspace will have 1/3 of the capacity it has now, or will each workspace created with the pipeline have the equivalent of a full A3?

Thanks in advance.

[Nimrod_Shalit]

@sapientcompany deployment pipelines works with embedded SKUs as well.

Pipelines are comprised of workspaces, that behave like any other workspace. You can choose to assign them to separate capacities, ot have them on the same one. when they share the capacity, it behaves as any other capacity with multiple workspaces. 

[sapientcompany]

Thank you.  So it is shared but do you know if a copy of the data is made for a dataset or if it is just metadata that gets copied?  My concern is since it is shared then I will consume resources by having 3 copies of a large dataset.

[Nimrod_Shalit · Accepted solution]

Each dataset stands on its own, so it will require data to be able to work in each of the stages.

Your problem is something multiple customers are facing, and it is usually solved in 2 ways:

1. Have a separate DB for each stage, so that dev/ test are loaded only with small amount of data. You can configure rules in a pipeline to control the connections across stages, so Prod is unaffected.
2. If you only have 1 data source to retrieve the data, you can use M parameters to control the amount of data that's being pulled in dev/test. Through param rules you can control the values across stages and make sure Prod stage is unaffected.