Date: Tue, 14 May 2013 17:23:57 -0700 From: Jeremy Chadwick <jdc@koitsu.org> To: Daniel Kalchev <daniel@digsys.bg> Cc: "freebsd-fs@freebsd.org" <freebsd-fs@freebsd.org> Subject: Re: NFS Performance issue against NetApp Message-ID: <20130515002357.GA21753@icarus.home.lan> In-Reply-To: <12D600DE-CBAB-40C6-B166-083DE7018E7E@digsys.bg> References: <1966772823.291493.1368362883964.JavaMail.root@erie.cs.uoguelph.ca> <5190335D.9090105@hub.org> <20130513005858.GA73875@icarus.home.lan> <94661399-66AC-4E83-B39B-0426442BB84C@hub.org> <12D600DE-CBAB-40C6-B166-083DE7018E7E@digsys.bg>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, May 15, 2013 at 03:19:21AM +0300, Daniel Kalchev wrote: > > > > >> Probably off-topic but worth pointing out: I do not know about Solaris, > >> but Linux has multiple layers of caching, and is well-known for doing > >> things like caching (and aggregating!) reads/writes to **block** devices > >> (this is why on Linux you have to make sure to avoid caching your > >> application use O_DIRECT with open(2) or other mechanisms -- the BSDs do > >> not do this, block devices are always non-cached). > > > > Caching *should* only come into play after the first run of the application … the first run after a reboot of the server shouldn't have anything in cache yet for caching to come into play > > > > Or, instead of issuing 30 separate NFS calls over the network, issue just one. With more latency the difference will be more pronounced. > > I believe Jeremy was referring more to the aggregating aspect, which might produce significant difference for poorly written software. Thanks Daniel -- yes, correct. :-) -- | Jeremy Chadwick jdc@koitsu.org | | UNIX Systems Administrator http://jdc.koitsu.org/ | | Mountain View, CA, US | | Making life hard for others since 1977. PGP 4BD6C0CB |
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130515002357.GA21753>