Date: Wed, 16 May 2012 07:40:02 +0200 (CEST) From: Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl> To: David Brodbeck <brodbd@uw.edu> Cc: freebsd-hackers@freebsd.org, Peter Jeremy <peterjeremy@acm.org> Subject: Re: NFS - slow Message-ID: <alpine.BSF.2.00.1205160737480.26713@wojtek.tensor.gdynia.pl> In-Reply-To: <CAHHaOuZQiOYpAEdhS0xqrN-KwOvj-7Rm4Vd5B_yKqxGKRamM0Q@mail.gmail.com> References: <alpine.BSF.2.00.1204272201440.6369@wojtek.tensor.gdynia.pl> <20120430085748.GA56921@server.vk2pj.dyndns.org> <alpine.BSF.2.00.1205010700240.5909@wojtek.tensor.gdynia.pl> <CAHHaOuZQiOYpAEdhS0xqrN-KwOvj-7Rm4Vd5B_yKqxGKRamM0Q@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>> same. am i doing something wrong? > > I found NFSv4 to be much *slower* than NFSv3 on FreeBSD, when I > benchmarked it a year or so ago. both are just right in you read (NFSv4 taking a bit more CPU), and both are awful at writes. for me now the only way to get NFS working well is to use unfsd with fsync commented out in sources. get very fast and not compliant with NFS protocol, which i don't care. i don't have database logs on NFS, but sometimes need to run non freebsd device without HDD and compile something using NFS disk.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.1205160737480.26713>