Re: Postgres performance comments from a MySQL user

Ernest E Vogelsinger <ernest@vogelsinger.at> writes:
> At 02:04 17.06.2003, Tom Lane said:
>> It's really hard to believe that you could see that kind of ratio from
>> any sort of cache effects, even kernel disk buffer cache which is
>> normally pretty large.

> I have EXPLAIN ANALYZE VERBOSE available.

That's the 3-sec-vs-10-msec case though.  I can easily believe that that
represents kernel disk caching effects --- that is, 3 sec is what it
really takes to read all the data from disk, but once it's in RAM you
can do the calculations in 10 msec.  The number that got my attention
was 20 minutes.  I don't see where that could come from, given the same
query plan and no change in other system load.

            regards, tom lane