Gian Merlino

> Any reason not to just allow requests to contain SET statements to be included prior to a insert/replace/select statement? That would work too, and would solve the "need to...

I think `SET` would make sense if our stance is that each HTTP call is its own session. I am not sure if other databases do this, so maybe it's...

> I don't know about it, could u show me an exmaple. But I agree that we can define a configuration to enable/disable the validation behaviour. In our deployment, we...

> @clintropolis now that segments are being cached on-demand on historicals (not all pre-loaded), and intra-AZ requests are much faster than S3 reads (intra-AZ requests are typically XXXµs, whereas S3...

IMO, rather than expanding the `ARRAY` functions to accept various kinds of scalars in array context, it's better to introduce a new function that explicitly accepts a scalar + an...

I'm wondering if you have a way to reproduce this on a fresh install? Like a dataset that can be loaded, where we can run this query on that dataset?...

> @gianm did you get a chance to review this? @djiangc I just fetched the zip file and loaded up the data to try to figure out what is going...

Hi @djiangc— thank you for the details! I tried this on Druid 26 or latest master, and got the following results in both the native engine & MSQ. It looks...

Sounds reasonable to me in general. I do believe @clintropolis is right that we would want to have a memory model where memory is associated with the query rather than...

> Want to create a way to partition the result set of a query (from either realtime/historical data nodes) into cacheable granular intervals that can either pulled from cache and...