Hello, question about timeseries vs timeboundary query performance. I have a table containing emails and customer IDs. I want to get the oldest time for an email and the oldest time for the same email but filtered on a customer ID. My app will have to do 2 timeboundary minTime queries (in parallel) to get those results, they should be fast because they're using the metadata. Whereas I can get the 2 results in only one timeseries query, but it should be slower ? Am I correct ? Which strategy is the fastest ?

Would inline filters help? Ie, something along the lines of

select latest(email) as latestOverall, latest(email) filter (where customerId='123') as latestForCust123