Date: Sat, 17 Feb 2007 14:34:11 -0600 From: Kirk Strauser <kirk@strauser.com> To: freebsd-questions@freebsd.org Subject: A bottleneck in gstripe? Message-ID: <200702171434.15077.kirk@strauser.com>
next in thread | raw e-mail | index | archive | help
--nextPart27907994.Xo71PBWzrl Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline I built a gstripe volume with 4 drives and a 128KB stripesize. When runnin= g=20 one particular application, gstat reports that stripe/stripe1 is 99% busy,= =20 although its four drives are running at less than 30% each. Am I=20 misinterpreting the numbers - maybe the total is the sum of the drives? -=20 or is there some giant overhead that I'm missing? I've set kern.geom.stripe.fast=3D1, and kern.geom.stripe.fast_failed stays = at=20 0. I don't have enough experience with geom_strip to even know where to go= =20 from here. Are stripe sizes likely to make much of a difference when the=20 heaviest load is when PostgreSQL is receiving massive imports? This is a=20 production system and I don't have the opportunity to play with it as much= =20 as I'd like, so any pointers to experiments likely to make a difference=20 would be most welcome. =2D-=20 Kirk Strauser --nextPart27907994.Xo71PBWzrl Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- iD8DBQBF12bH5sRg+Y0CpvERAi18AKCC/t+h1h7FoEZtKmADCWjjEyWDGgCfcghd gL6arVFJzsGbZkHdE8h0W6s= =MVCY -----END PGP SIGNATURE----- --nextPart27907994.Xo71PBWzrl--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200702171434.15077.kirk>