Downsizing a Dataset

Is there a way to reduce the number of rows in a dataset by deleting rows prior to a given date? My dataset has some old, irrelevant data in it that can be purged. I am seeing in the documentation that the deletion and recreation of a dataset will break any existing Analyses, so that is not an option for me.

Thanks!

You can edit the dataset and add a filter to only get data that is after your chosen data. Once you save, any analyses and dashboards using this dataset will see the new data. This will not break existing analyses because you are updating the existing dataset and not deleting and creating a new one.

What about the cases where you have a static dataset that has a ROLLING 2 years worth of history. Because of SPICE storage costs, how would you go about deleting older data in that dataset? Do you have to do a full refresh?

Yes you would need to do a full refresh in those cases. But actually having ttl expiration could be a good feature and would complement incremental ingestion quite nicely.