[HACKERS] what is the smallest working page size for postgresql
Hi Could anybody tell me what is the smallest working page size for postgresql ? I have a table where access is highly random over huge table getting usually only one small tuple from each page. One way to get more performance could be using smaller page size, so the per-tuple read overhead would be smaller. Would 4k pages work ? what about 2k and 1k ? 512bytes ? What would it take, to make only heap pages small and keep index pages larger ? Probably at least per-tablespace or per-pagesize split shared buffer space and changes in caching algorithms ? Has anyone tested if 8k is big enough to trigger (in my case unneccesary) read-ahead on disks/controllers/devices/filesystem ? --- Hannu ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] what is the smallest working page size for postgresql
Hannu Krosing [EMAIL PROTECTED] wrote Could anybody tell me what is the smallest working page size for postgresql ? I have a table where access is highly random over huge table getting usually only one small tuple from each page. One way to get more performance could be using smaller page size, so the per-tuple read overhead would be smaller. Would 4k pages work ? what about 2k and 1k ? 512bytes ? What would it take, to make only heap pages small and keep index pages larger ? Probably at least per-tablespace or per-pagesize split shared buffer space and changes in caching algorithms ? I recall there was a discussion several weeks ago: http://archives.postgresql.org/pgsql-performance/2005-12/msg00120.php I bet block size less than 512 won't bring you any benefits, since that's the physical disk sector size limit. Regards, Qingqing ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
