Date: Thu, 10 Feb 2005 18:15:06 -0500 From: Ean Kingston <ean@hedron.org> To: freebsd-questions@freebsd.org Subject: Re: FreeBSD 5.3 MySQL Performance Message-ID: <200502101815.07092.ean@hedron.org> In-Reply-To: <420BD7DD.5000304@fastclick.com> References: <20050209162202.I31921@knight.ixsystems.net> <1108005975.683.47.camel@zircon.zircon.seattle.wa.us> <420BD7DD.5000304@fastclick.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On February 10, 2005 04:53 pm, Jeff Behl wrote: > >Any so-called "benchmark" comparing Linux to anything else (especially > >windoze) has been polluted by the tradition in the linux/windoze world > >of running their disks in the completely unsafe "asynchronous" mode so > >popular with the ATA disk drive manufacturers. This method means that > >you never actually know whether or not the drive ever writes your data > >on the disk. It could just sit in the cache waiting for a power failure > >so that you lose everything. This "async" mode means that the > >benchmarks "look" fast but are completely unsafe. > > so by this logic, if i re-mount my partitions async i can get the same > performance? this isn't meant as a rub, i would seriously consider > doing this if it were the case. i'd like to know any and all ways i can > make mysql faster. we have fleats of mysql servers with redundant > data. the loss of a server due to corruption is not problematic You will get significant speed increase out of your filesystem(s) if you mount them async. BUT if you don't unmount them properly you will have corrupted filesystems. I do this with /tmp. -- Ean Kingston E-Mail: ean AT hedron DOT org URL: http://www.hedron.org/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200502101815.07092.ean>