Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread sparikh
I will try to request both perf and strace to be installed. But I am > not quite sure as the VMs are managed by third party. Will keep you > posted... What do you mean by VM? Is this a virtualized environment or bare hardware? Yes, they are virtualized environments. Sorry about the confusi

Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread Tomas Vondra
On 22.10.2013 23:50, sparikh wrote: > Sorry, it was typo from my side. I meant strace only. OK. > > I will try to request both perf and strace to be installed. But I am > not quite sure as the VMs are managed by third party. Will keep you > posted... What do you mean by VM? Is this a virtualize

Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread Tomas Vondra
On 22.10.2013 23:41, sparikh wrote: >>From Primary: > > relname relpages > pg_toast_176731812819 > pg_toast_17594161660 > pg_toast_17972121902 > pg_toast_1758777190 > pg_toast_1853729108 > pg_toast_1757826638 > pg_toast_17673_index 19984 > pg_

Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread sparikh
Sorry, it was typo from my side. I meant strace only. I will try to request both perf and strace to be installed. But I am not quite sure as the VMs are managed by third party. Will keep you posted... The main thing puzzling to me is Explain Plan with Analyze takes couple of secs to execute the o

Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread sparikh
>From Primary: relname relpages pg_toast_17673 1812819 pg_toast_17594 161660 pg_toast_17972 121902 pg_toast_17587 77190 pg_toast_18537 29108 pg_toast_17578 26638 pg_toast_17673_index19984 pg_toast_17868 14911 pg_toast_17594_index2208 pg_toast_10722461922 pg_toast_17587_index

[PERFORM] Logic of lowering seq_page_cost for SSD?

2013-10-22 Thread Josh Berkus
All, I've often seen people lower seq_page_cost for SSD access. This has the effect of raising CPU costs relative to the costs of disk access. However, CPUs have also gotten lots faster, so I'm not sure that results in a better cost balance. Has anyone done extensive testing on this? What did y

Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread Tomas Vondra
On 22.10.2013 02:00, sparikh wrote: >> Do you suggest if I remove all the data files from /data/base folder >> of standby and again rebuild using rsync from primary ? do you see >> any issues there.? This is just to rule out any fragmentation on >> standby side. > > The EXPLAIN really should n

Re: [PERFORM] Hot Standby performance issue

2013-10-22 Thread Tomas Vondra
On 22.10.2013 06:49, Magnus Hagander wrote: > > On Oct 22, 2013 1:14 AM, "Tomas Vondra" > wrote: >> >> On 22.10.2013 00:59, sparikh wrote: >> > Yes, Expalin without Analyze is taking long. It is weird. In the >> > pg_stat_activity Explain was the only query running. So serve