On Fri, Jan 2, 2009 at 20:30, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> One point that nobody seems to have focused on is whether Alex's
> less-compressed table is faster or slower to access than the original.
> I dunno if he has any easy way of investigating that for his typical
> query mix, but it's certainly a fair question to ask.
Other than the quick pgbench numbers I posted upthread where 8.4 blew
8.3 out of the water with a substring. Not really, this table is
mainly insert. A few times a day everything inserted that day gets
selected. So while I'm almost positive 8.4 is faster, its probably
not really noticeable in my workload. That being said here are some
quick numbers:
(see attached q.sql for how uninteresting the query is, also this is
so slow mainly due to the lack of it using an index, it seq-scans the
entire table :()
./pgbench -T600 -n -f q.sql
8.4 with 8.3 TOAST: 6.250298
8.4: 6.460312
(note I dont actually use substring on this table...)
./pgbench -T60 -n -f substring.sql
8.4 w 8.3 TOAST: 12.613394
8.4: 6347.456596