Palle Girgensohn <girgen@pingpong.net> writes:
>> No surprise. vacuum analyze produces an exact total row count, whereas
>> analyze can only produce an approximate total row count (since it only
>> samples the table rather than groveling over every row). Sometimes the
>> approximate count will be far enough off to affect the estimates.
> Reasonable. Thanks for clarifying that. In the normal case, vacuum analyze
> is better, I guess?
If you intend to do both steps, the combined command is definitely
better than issuing them separately. I wouldn't say that you need to
do the combined command in situations where you'd otherwise do just
one.
regards, tom lane