On tis, 2012-03-13 at 15:44 -0400, Tom Lane wrote:
> I wonder whether it'd be worth recommending that people do an initial
> ANALYZE with a low stats target, just to get some stats in place,
> and then go back to analyze at whatever their normal setting is.
Perhaps going even further, ANALYZE could have a quick-and-dirty mode,
where it just analyzes say 10 random pages per table and fills in as
much data as it can from that. And then it does the real analyze. (Or
it could even insert fake statistics to trigger autoanalyze.) That way,
you could have a database back in business in less than a minute.