Re: [PERFORM] Cost of creating an emply WAL segment

2011-06-24 Thread Greg Smith
On 06/24/2011 02:55 PM, Shaun Thomas wrote: On 06/24/2011 11:18 AM, Greg Smith wrote: sync=14525.296 s, total=14786.868 s What!? 6% of 8GB is just shy of 500MB. That's not a small amount, exactly, but it took 14525 seconds to call syncs for those writes? What kind of ridiculous IO w

Re: [PERFORM] Cost of creating an emply WAL segment

2011-06-24 Thread Shaun Thomas
On 06/24/2011 11:18 AM, Greg Smith wrote: sync=14525.296 s, total=14786.868 s What!? 6% of 8GB is just shy of 500MB. That's not a small amount, exactly, but it took 14525 seconds to call syncs for those writes? What kind of ridiculous IO would cause something like that? That's even

Re: [PERFORM] Cost of creating an emply WAL segment

2011-06-24 Thread Greg Smith
On 06/24/2011 10:43 AM, Devrim GÜNDÜZ wrote: Has anyone measured the cost of creating empty WAL segments while the database is running? Actually, when is the new file created? Just after one segment is filled up, or some time before then? What happens during WAL segment creation? If there are pe

[PERFORM] Cost of creating an emply WAL segment

2011-06-24 Thread Devrim GÜNDÜZ
Hi, Has anyone measured the cost of creating empty WAL segments while the database is running? Actually, when is the new file created? Just after one segment is filled up, or some time before then? What happens during WAL segment creation? If there are pending transactions to be committed, do w