Closed bertsky closed 6 months ago
Currently we build minimum, medium, maximum and maximum-cuda in parallel.
Except for the latter, this offers no time advantage, just consumes more resources (credits).
I suggest we change to
and have the respective jobs depend on each other in the CI.
The latest Docker build fails with a timeout. Shouldn't the maximum builds now be faster than before because they only have to build on top of medium?
Currently we build minimum, medium, maximum and maximum-cuda in parallel.
Except for the latter, this offers no time advantage, just consumes more resources (credits).
I suggest we change to
and have the respective jobs depend on each other in the CI.