Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 26 Apr 2006 07:05:23 -0500
From:      Jonathan Horne <jhorne@dfwlp.com>
To:        freebsd-questions@freebsd.org
Subject:   Re: How to verify speed of a 1Gb/s network?
Message-ID:  <200604260705.23623.jhorne@dfwlp.com>
In-Reply-To: <274988B5-AE3C-4F97-A811-E9BC1F43A580@shire.net>
References:  <20060426031606.33136.qmail@web33302.mail.mud.yahoo.com> <274988B5-AE3C-4F97-A811-E9BC1F43A580@shire.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday 25 April 2006 23:17, Chad Leigh -- Shire.Net LLC wrote:
> On Apr 25, 2006, at 9:16 PM, Rob wrote:
> > Hi,
> >
> > How can I verify that a 1Gb/s network is indeed
> > operating at its optimal speed? I tried this:
> >
> > [master]$ ping -s 65507 node
> > 65515 bytes from node: icmp_seq=0 ttl=64 time=1.97 ms
> > 65515 bytes from node: icmp_seq=1 ttl=64 time=1.95 ms
> > 65515 bytes from node: icmp_seq=2 ttl=64 time=1.94 ms
> > 65515 bytes from node: icmp_seq=3 ttl=64 time=1.97 ms
> >
> > (I tried many times, over a long period of time
> > to get these typical values).
> >
> >> From this I conclude that it takes about 1.95 ms
> >
> > for 65515 x 8 bits to go forth and back between
> > master and node.
> >
> > Ideally, on a 1Gbit/s network, the time should be:
> >   65515 x 8 x 2 / (10243) = 0.98 ms
> > (x 2 for the roundtrip signal forth and back
> > and 10243 is the 1G of the network)
> >
> > May I now conclude that the real-time is about
> > two times the ideal-time? I wonder if this indicates
> > a problem of the network?
> > And is this a proper test of this Gbit/s network?
> >
> > Thanks,
> > Rob.
> >
> > PS: I verified my calculation method for two
> > computers here on a 100Mbit/s network, from which
> > I get:
> >    time with ping: 12.4 ms
> >    ideal calculated time: 10 ms
> > which is an acceptable difference
>
> I would suspect that a ping is not a valid test as it does not test
> throughput and the send and reception phases have a large influence
> on the out come.  Ie, the time for the send and reception to take
> place is long enough compared to the fast network that the results
> are skewed.  Try an ftp or other non-encrypted data transfer with a
> large enough file that the startup and wind-down won't affect and
> skew it.  Probably still not a definitive test
>
> btw, here is a test of my gbit network using your ping test
>
> 15 packets transmitted, 15 packets received, 0% packet loss
> round-trip min/avg/max/stddev = 0.128/0.227/0.342/0.061 ms

a few years back, i had a gigabit fiber switch, and 2 intel gigabit fiber 
cards that i put in my 2 fastest computers (at the time, dual p3 1000 and 
dual p3 933).  they both had 10k rpm ultra160 SCSI drives.  the fastest i 
could get for continuous transfer (i made some gigantic zip files containing 
several .iso files) was about 250mbit.

jonathan



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200604260705.23623.jhorne>