Date: Mon, 2 Aug 2010 07:44:20 -0500 From: David Kelly <dkelly@hiwaay.net> To: Corey Smith <corsmith@gmail.com> Cc: freebsd-questions@freebsd.org, "Jason C. Wells" <jcw@speakeasy.net> Subject: Re: Typical Network Performance Message-ID: <5628C9CD-0F16-4C0E-8B89-B4ECCA35C933@hiwaay.net> In-Reply-To: <AANLkTi=g%2BBGLJRQfyz7v3dSQ6k%2BxNQzVEEnSBdxpJfGF@mail.gmail.com> References: <4C55E4B5.7000201@speakeasy.net> <8627B125-F3BB-42B2-98CF-600E21A93A2D@hiwaay.net> <AANLkTi=g%2BBGLJRQfyz7v3dSQ6k%2BxNQzVEEnSBdxpJfGF@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Aug 1, 2010, at 11:31 PM, Corey Smith wrote: > On Sun, Aug 1, 2010 at 7:30 PM, David Kelly <dkelly@hiwaay.net> wrote: >> Gigabit ethernet from a 2.8 GHz P4 to or from MacPro I am only = limited by disk data rate. About 60 MB/sec on one end of the disk, more = on the other end of the disk. >=20 > Did you try realtime monitoring your network interface? >=20 > # route -n get <remoteip> > interface: <yourinterface> >=20 > # netstat -I <yourinterface> -w 1 No. I saw numbers that I was reasonably happy with and didn't pursue = further. > Do you see errors on the interface? Nope. 60 MB/sec via FTP is about 60% of gigabit and was faster than some = disk accesses. > # netstat -I <yourinterface> >=20 > Another trick to eliminate disk io from the equation is to use nc: >=20 > machine1 <freebsd>: > # nc -o -l 2000 > /dev/null >=20 > machine2: > # dd if=3D/dev/zero bs=3D1M count=3D50 | nc machine1 2000 60 MB/sec was the average over gigabytes of data. Real data. Real = network wire. -- David Kelly N4HHE, dkelly@HiWAAY.net =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Whom computers would destroy, they must first drive mad.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5628C9CD-0F16-4C0E-8B89-B4ECCA35C933>