Date: Wed, 15 May 2013 03:19:21 +0300 From: Daniel Kalchev <daniel@digsys.bg> To: "Marc G. Fournier" <scrappy@hub.org> Cc: "freebsd-fs@freebsd.org" <freebsd-fs@freebsd.org> Subject: Re: NFS Performance issue against NetApp Message-ID: <12D600DE-CBAB-40C6-B166-083DE7018E7E@digsys.bg> In-Reply-To: <94661399-66AC-4E83-B39B-0426442BB84C@hub.org> 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>
next in thread | previous in thread | raw e-mail | index | archive | help
>=20 >> 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). >=20 > Caching *should* only come into play after the first run of the applicatio= n =E2=80=A6 the first run after a reboot of the server shouldn't have anythi= ng in cache yet for caching to come into play=20 >=20 Or, instead of issuing 30 separate NFS calls over the network, issue just on= e. With more latency the difference will be more pronounced. I believe Jeremy was referring more to the aggregating aspect, which might p= roduce significant difference for poorly written software.=20 Daniel=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?12D600DE-CBAB-40C6-B166-083DE7018E7E>