Re: [PERFORM] CPU spikes and transactions

2014-05-14 Thread Jeff Janes
On Tue, May 13, 2014 at 4:04 PM, Dave Owens wrote: > Hi, > > Apologies for resurrecting this old thread, but it seems like this is > better than starting a new conversation. > > We are now running 9.1.13 and have doubled the CPU and memory. So 2x 16 > Opteron 6276 (32 cores total), and 64GB memo

Re: [PERFORM] CPU spikes and transactions

2014-05-14 Thread Merlin Moncure
On Tue, May 13, 2014 at 6:04 PM, Dave Owens wrote: > Hi, > > Apologies for resurrecting this old thread, but it seems like this is > better than starting a new conversation. > > We are now running 9.1.13 and have doubled the CPU and memory. So 2x 16 > Opteron 6276 (32 cores total), and 64GB memo

Re: [PERFORM] CPU spikes and transactions

2014-05-13 Thread Dave Owens
Hi, Apologies for resurrecting this old thread, but it seems like this is better than starting a new conversation. We are now running 9.1.13 and have doubled the CPU and memory. So 2x 16 Opteron 6276 (32 cores total), and 64GB memory. shared_buffers set to 20G, effective_cache_size set to 40GB.

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Merlin Moncure
On Tue, Oct 15, 2013 at 12:26 PM, Julien Cigar wrote: > On Tue, Oct 15, 2013 at 08:59:08AM -0700, Tony Kay wrote: >> On Mon, Oct 14, 2013 at 4:42 PM, Tomas Vondra wrote: >> >> > On 15.10.2013 01:00, Tony Kay wrote: >> > > Hi, >> > > >> > > I'm running 9.1.6 w/22GB shared buffers, and 32GB overall

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Tomas Vondra
On 15.10.2013 21:07, Tony Kay wrote: > > On Tue, Oct 15, 2013 at 10:26 AM, Julien Cigar > wrote: > > > for two reasons: > > - some of the overhead of bgwriter and checkpoints is more or less > linear in the size of shared_buffers, for example it could be > possible t

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Julien Cigar
On Tue, Oct 15, 2013 at 12:07:38PM -0700, Tony Kay wrote: > On Tue, Oct 15, 2013 at 10:26 AM, Julien Cigar wrote: > > > > > for two reasons: > > > > - some of the overhead of bgwriter and checkpoints is more or less linear > > in the size of shared_buffers, for example it could be possible that a

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Tony Kay
On Tue, Oct 15, 2013 at 10:26 AM, Julien Cigar wrote: > > for two reasons: > > - some of the overhead of bgwriter and checkpoints is more or less linear > in the size of shared_buffers, for example it could be possible that a > large quantity of data could be dirty when a checkpoint occurs). > >

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Julien Cigar
On Tue, Oct 15, 2013 at 08:59:08AM -0700, Tony Kay wrote: > On Mon, Oct 14, 2013 at 4:42 PM, Tomas Vondra wrote: > > > On 15.10.2013 01:00, Tony Kay wrote: > > > Hi, > > > > > > I'm running 9.1.6 w/22GB shared buffers, and 32GB overall RAM on a > > > 16 Opteron 6276 CPU box. We limit connections

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Tony Kay
Thanks for the tip. I forgot there were kernel stats on spinlocks. I'm not sure we'll be able to get it to tip in a test environment, and we're unwilling to revert the code in production in order to have our users trigger it. We'll try triggering it on our test server, and if we manage, I'll get y

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Tony Kay
On Mon, Oct 14, 2013 at 4:42 PM, Tomas Vondra wrote: > On 15.10.2013 01:00, Tony Kay wrote: > > Hi, > > > > I'm running 9.1.6 w/22GB shared buffers, and 32GB overall RAM on a > > 16 Opteron 6276 CPU box. We limit connections to roughly 120, but > > our webapp is configured to allocate a thread-lo

Re: [PERFORM] CPU spikes and transactions

2013-10-15 Thread Merlin Moncure
On Mon, Oct 14, 2013 at 6:45 PM, Tomas Vondra wrote: > On 15.10.2013 01:26, Tony Kay wrote: >> Hi Calvin, >> >> Yes, I have sar data on all systems going back for years. >> >> Since others are going to probably want to be assured I am really >> "reading the data" right: >> >> - This is 92% user CP

Re: [PERFORM] CPU spikes and transactions

2013-10-14 Thread Tomas Vondra
On 15.10.2013 01:26, Tony Kay wrote: > Hi Calvin, > > Yes, I have sar data on all systems going back for years. > > Since others are going to probably want to be assured I am really > "reading the data" right: > > - This is 92% user CPU time, 5% sys, and 1% soft > - On some of the problems, I _

Re: [PERFORM] CPU spikes and transactions

2013-10-14 Thread Tomas Vondra
On 15.10.2013 01:00, Tony Kay wrote: > Hi, > > I'm running 9.1.6 w/22GB shared buffers, and 32GB overall RAM on a > 16 Opteron 6276 CPU box. We limit connections to roughly 120, but > our webapp is configured to allocate a thread-local connection, so > those connections are rarely doing anything m

Re: [PERFORM] CPU spikes and transactions

2013-10-14 Thread Tony Kay
Hi Calvin, Yes, I have sar data on all systems going back for years. Since others are going to probably want to be assured I am really "reading the data" right: - This is 92% user CPU time, 5% sys, and 1% soft - On some of the problems, I _do_ see a short spike of pgswpout's (memory pressure), b

[PERFORM] CPU spikes and transactions

2013-10-14 Thread Tony Kay
Hi, I'm running 9.1.6 w/22GB shared buffers, and 32GB overall RAM on a 16 Opteron 6276 CPU box. We limit connections to roughly 120, but our webapp is configured to allocate a thread-local connection, so those connections are rarely doing anything more than half the time. We have been running smo