You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think once you get into that you might be better off using postgres with the hstore or json column types; they're probably going to be way more optimized than what we're doing. Although there might be interesting opportunities for compression given the time-series format!
I'd be down for adding strings though; I think that could come in handy for stuff like logging and a few people have asked for it. It does probably do funny things to the bucketing approach that is really why the compression works in the first place so it's not completely trivial; although you could do a trivial implementation using the protobufs.
Currently only doubles are allowed as value types. It would be great if you could pass in Python longs or even dicts.
The text was updated successfully, but these errors were encountered: