Evgeny Gridasov <eugrid@xxxxxxxxxxxx> writes: > Running the same query on pg 8.2 through EXPLAIN ANALYZE takes 4x-10x time as running it without it. If your machine has slow gettimeofday() this is not surprising. 8.2 is no worse (or better) than any prior version. Some quick arithmetic from your results suggests that gettimeofday() is taking about 3.3 microseconds, which is indeed pretty awful. What sort of machine is this exactly? regards, tom lane
