There have been a variety of jobs running the DAS e2e tests on different releases i.e. 4.18, 4.19, 4.20.[1]

All these jobs are pre-merges and there could be more in future when we support the next versions of OCP.

Though its a good practice to test the PR changes via pre-merges on all the versions, we have been hitting the bottleneck while provisioning the OCP clusters with GPU attached to test the feature and hitting the "Quota Exceeded" failures[2]

So, the idea is to keep one pre-merge job i.e. the latest version of OCP that supports DAS(NVIDIA GPU operator) and convert/modify all the other existing jobs to be periodic(may be run a job in a day or two?)

[1] - https://github.com/openshift/release/blob/master/ci-operator/config/openshift/instaslice-operator/openshift-instaslice-operator-next.yaml 
[2] - https://github.com/openshift/release/pull/71889#issuecomment-3729496953