On Mon, 2006-12-18 at 14:47 +0900, Takayuki Tsunakawa wrote: > Hello, Itagaki-san, all > > Sorry for my long mail. I've had trouble in sending this mail because > it's too long for pgsql-hackers to accept (I couldn't find how large > mail is accepted.) So I'm trying to send several times. > Please see the attachment for the content.
Your results for fsync are interesting. I've noticed that a checkpoint seems to increase the activity on the WAL drive as well as increasing I/O wait times. That doesn't correspond to any real increase in WAL traffic I'm aware of. Have you tried setting deadline scheduler on the WAL device and CFQ on the data device? That should allow the I/Os to move through different queues and prevent interference. -- Simon Riggs EnterpriseDB http://www.enterprisedb.com ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster