no partial (or other) index affecting these results. just a trigger the outcome of which was affected by the value of that particular boolean. sorry, should have looked more carefully... > -----Original Message----- > From: Tom Lane [mailto:tgl@xxxxxxxxxxxxx] > Sent: Tuesday, May 23, 2006 4:00 PM > To: George Pavlov > Cc: pgsql-general@xxxxxxxxxxxxxx > Subject: Re: [GENERAL] assymetry updating a boolean (=FALSE > faster than =TRUE) > > "George Pavlov" <gpavlov@xxxxxxxxxxxxxx> writes: > > Here is something that seems anomalous to me: when I set a > boolean field > > to FALSE performance is much better than when I set it to TRUE. Any > > reason for FALSE to be favored over TRUE? > > It isn't. You're measuring something else than you think you're > measuring ... maybe a partial index? Varying effects of where the > free space is in the table? In a test case I see no significant > difference: > > regression=# create table foo(f1 int, f2 bool); > CREATE TABLE > regression=# insert into foo select x, true from > generate_series(1,100000) x; > INSERT 0 100000 > regression=# \timing > Timing is on. > regression=# vacuum foo; > VACUUM > Time: 318.153 ms > regression=# update foo set f2 = true; > UPDATE 100000 > Time: 2553.014 ms > regression=# vacuum foo; > VACUUM > Time: 410.356 ms > regression=# update foo set f2 = false; > UPDATE 100000 > Time: 2453.620 ms > regression=# vacuum foo; > VACUUM > Time: 405.955 ms > regression=# update foo set f2 = true; > UPDATE 100000 > Time: 2497.413 ms > regression=# vacuum foo; > VACUUM > Time: 408.403 ms > regression=# update foo set f2 = false; > UPDATE 100000 > Time: 2458.824 ms > regression=# > > regards, tom lane >
