Mailing List Archive: NANOG: users

Speed Testing and Throughput testing

Index | Next | Previous | View Threaded

mark.urbach at pnpt

Nov 2, 2009, 1:56 PM

Post #1 of 20 (1193 views)
Permalink

Speed Testing and Throughput testing

Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?

Do you have a server/software that customer can test too?

Thanks,
Mark Urbach
PinPoint Communications, Inc.
100 N. 12th St Suite 500
Lincoln, NE 68508
402-438-6211 ext 1923 Office
402-660-7982 Cell
mark.urbach [at] pnpt
[cid:image003.jpg [at] 01CA5BD5]

Attachments:

image003.jpg (1.93 KB)

jack at crepinc

Nov 2, 2009, 2:51 PM

Post #2 of 20 (1162 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

iperf is fairly standard and supports some handy features -
http://en.wikipedia.org/wiki/Iperf

-Jack Carrozzo

On Mon, Nov 2, 2009 at 4:56 PM, Mark Urbach <mark.urbach [at] pnpt> wrote:
> Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?
>
> Do you have a server/software that customer can test too?
>
>
>
> Thanks,
> Mark Urbach
> PinPoint Communications, Inc.
> 100 N. 12th St Suite 500
> Lincoln, NE 68508
> 402-438-6211 ext 1923 Office
> 402-660-7982 Cell
> mark.urbach [at] pnpt
> [cid:image003.jpg [at] 01CA5BD5]
>
>

nanog at daork

Nov 2, 2009, 2:54 PM

Post #3 of 20 (1167 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

On 3/11/2009, at 10:56 AM, Mark Urbach wrote:

> Anyone have a good solution to get "accurate" speed results when
> testing at 10/100/1000 Ethernet speeds?

If you want accuracy, you want to buy a packet generator/router tester
unit.

I just built a tool for a customer (a last-mile network provider) that
runs a series of iperf tests over several days, and generates a report.
iperf works well enough, but it seems to be much better when driven by
humans, vs. driven by scripts.

I'm not aware of any free tools that do just ethernet frames.

> Do you have a server/software that customer can test too?

Not sure what you're after here - do you want to host your own
speedtest.net-like service so your customers can self-test their
access links? Does this mean much, or should they be testing against a
server outside your network?
Also, if you host your own service and you're talking about
10/100/1000mbit connections, you might want to put something in place
that prevents several people testing at once.

--
Nathan Ward

azher at hep

Nov 2, 2009, 3:09 PM

Post #4 of 20 (1166 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

perfsonar livecd offers npad service that remote hosts can connect and
see the performance and results.
http://www.internet2.edu/performance/toolkit/index.html

TcpOptimizer helps tunning the tcp/ip for windows systems.
http://www.speedguide.net/downloads.php

nuttcp is good to generate packets/sec.

-Azher

Nathan Ward wrote:
> On 3/11/2009, at 10:56 AM, Mark Urbach wrote:
>
>> Anyone have a good solution to get "accurate" speed results when
>> testing at 10/100/1000 Ethernet speeds?
>
> If you want accuracy, you want to buy a packet generator/router tester
> unit.
>
> I just built a tool for a customer (a last-mile network provider) that
> runs a series of iperf tests over several days, and generates a report.
> iperf works well enough, but it seems to be much better when driven by
> humans, vs. driven by scripts.
>
> I'm not aware of any free tools that do just ethernet frames.
>
>> Do you have a server/software that customer can test too?
>
> Not sure what you're after here - do you want to host your own
> speedtest.net-like service so your customers can self-test their
> access links? Does this mean much, or should they be testing against a
> server outside your network?Also, if you host your own service and
> you're talking about 10/100/1000mbit connections, you might want to
> put something in place that prevents several people testing at once.
>
> --
> Nathan Ward
>
>
-Azher

meekjt at gmail

Nov 2, 2009, 3:22 PM

Post #5 of 20 (1163 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

I use iperf with packet capture on both sides, then analyze the packet
capture for per-second throughput and re-transmits. I usually do 10
TCP streams for 30 seconds.

Note that on GigE with significant RTTs (5-15 ms) some TCP tuning is
needed to deal with the bandwidth delay product. It is also possible
that Ethernet drivers will have an effect. Local testing of the pair
of test machines should be done if you can't get to about 980 Mbps on
a Gig link (keeping in mind the comment about TCP tuning as latency
increases).

Jon

On Mon, Nov 2, 2009 at 4:56 PM, Mark Urbach <mark.urbach [at] pnpt> wrote:
> Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?
>
> Do you have a server/software that customer can test too?
>
>
>
> Thanks,
> Mark Urbach
> PinPoint Communications, Inc.
> 100 N. 12th St Suite 500
> Lincoln, NE 68508
> 402-438-6211 ext 1923 Office
> 402-660-7982 Cell
> mark.urbach [at] pnpt
> [cid:image003.jpg [at] 01CA5BD5]
>
>

andreas at naund

Nov 2, 2009, 3:27 PM

Post #6 of 20 (1161 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

Hello,

On Mon, Nov 02, 2009 at 03:56:56PM -0600, Mark Urbach wrote:
> Anyone have a good solution to get "accurate" speed results when
> testing at 10/100/1000 Ethernet speeds?

iperf. Check

http://www.nanog.org/meetings/nanog43/abstracts.php?pt=MjkmbmFub2c0Mw==&nm=nanog43

and

http://www.merit.edu/mail.archives/nanog/2009-03/threads.html#00388

> Do you have a server/software that customer can test too?

You can set up your own iperf server on your net, or install the
server side parts of the browser based software from 'speedtest.net'
http://speedtest.net/mini.php
Note: IMHO that one has issues selecting the correct test based on
the perceived RTT and will deliver misleading results sometime.

-andreas
--
Andreas Ott K6OTT andreas [at] naund

tvhawaii at shaka

Nov 2, 2009, 3:30 PM

Post #7 of 20 (1161 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

Nathan Ward wrote:
> On 3/11/2009, at 10:56 AM, Mark Urbach wrote:
>
>> Anyone have a good solution to get "accurate" speed results when
>> testing at 10/100/1000 Ethernet speeds?

An NDT server?... such as:
http://ndt.anl.gov:7123/

ras at e-gerbil

Nov 2, 2009, 4:02 PM

Post #8 of 20 (1162 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

On Mon, Nov 02, 2009 at 01:30:18PM -1000, Michael Painter wrote:
> Nathan Ward wrote:
> >On 3/11/2009, at 10:56 AM, Mark Urbach wrote:
> >
> >>Anyone have a good solution to get "accurate" speed results when
> >>testing at 10/100/1000 Ethernet speeds?
>
> An NDT server?... such as:
> http://ndt.anl.gov:7123/

I just tested that server, and couldn't get any results which were even
vaguely close to accurate. Of course it probably didn't help that the
only routes I could find to the test server were either Chicago - Palo
Alto - Chicago or Chicago - Ashburn - Chicago, but this doesn't seem
like it would ever be useful for testing gigabit anything.

For end user testing, I've actually seen reasonable results from
speedtest.net. http://www.speedtest.net/result/610596179.png for
example, better than ndt.anl.gov at any rate. :P

For quick and dirty high speed Internet testing up to a gigabit, this is
my favorite standby (it often helps to eliminate your local disk from
the equation by writing the downloaded file to /dev/null too):

> fetch -o /dev/null http://cachefly.cachefly.net/100mb.test
/dev/null 100% of 100 MB 102 MBps

But the best (and conveniently enough the most commonly used) tool for
in-depth high speed testing was already mentioned, iperf. Another useful
tool if you're trying to troubleshoot tcp issues is
http://www.tcptrace.org/.

--
Richard A Steenbergen <ras [at] e-gerbil> http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)

morrowc.lists at gmail

Nov 2, 2009, 6:00 PM

Post #9 of 20 (1152 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

On Mon, Nov 2, 2009 at 4:56 PM, Mark Urbach <mark.urbach [at] pnpt> wrote:
> Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?
>
> Do you have a server/software that customer can test too?

One wonders how netnod does this... I believe they put in some servers
specifically so their local users could verify that bw bought was bw
received...

Maybe someone from netnod even wrote up their
methods/procedures/process/utilities/tools? :) (Maybe one would even
give a talk about it at an upcoming meeting?)

-Chris

benoit.vannier at apog

Nov 3, 2009, 2:01 AM

Post #10 of 20 (1129 views)
Permalink

RE: Speed Testing and Throughput testing [In reply to]

Hello,

Iperf is pretty good at this ... It s free

Ben

-----Message d'origine-----
De : Mark Urbach [mailto:mark.urbach [at] pnpt]
Envoyé : lundi 2 novembre 2009 22:57
À : nanog [at] nanog
Objet : Speed Testing and Throughput testing

Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?

Do you have a server/software that customer can test too?

Thanks,
Mark Urbach
PinPoint Communications, Inc.
100 N. 12th St Suite 500
Lincoln, NE 68508
402-438-6211 ext 1923 Office
402-660-7982 Cell
mark.urbach [at] pnpt
[cid:image003.jpg [at] 01CA5BD5]

jason at biel-tech

Nov 3, 2009, 4:05 AM

Post #11 of 20 (1132 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

Please take note with using iperf that you'll want to make sure the
appropriate TCP Window Size has been negotiated. We recently did some
testing with systems that had decided to pick less than optimal window sizes
and in turn had to manually set the size within iperf options.

Jason

On Tue, Nov 3, 2009 at 4:01 AM, Benoit VANNIER <benoit.vannier [at] apog>wrote:

> Hello,
>
> Iperf is pretty good at this ... It s free
>
>
> Ben
>
>
> -----Message d'origine-----
> De : Mark Urbach [mailto:mark.urbach [at] pnpt]
> Envoyé : lundi 2 novembre 2009 22:57
> À : nanog [at] nanog
> Objet : Speed Testing and Throughput testing
>
> Anyone have a good solution to get "accurate" speed results when testing at
> 10/100/1000 Ethernet speeds?
>
> Do you have a server/software that customer can test too?
>
>
>
> Thanks,
> Mark Urbach
> PinPoint Communications, Inc.
> 100 N. 12th St Suite 500
> Lincoln, NE 68508
> 402-438-6211 ext 1923 Office
> 402-660-7982 Cell
> mark.urbach [at] pnpt
> [cid:image003.jpg [at] 01CA5BD5]
>
>
>

--
Jason Biel

bclark at spectraaccess

Nov 3, 2009, 4:18 AM

Post #12 of 20 (1130 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

True, we usually find Linux based machines work better running IPerf
then Windows (at least out of the box) because of the TCP window
size....well Windows XP at least, don't know about Vista or 7.
Jason Biel wrote:

Please take note with using iperf that you'll want to make sure the
appropriate TCP Window Size has been negotiated. We recently did some
testing with systems that had decided to pick less than optimal window sizes
and in turn had to manually set the size within iperf options.

Jason

On Tue, Nov 3, 2009 at 4:01 AM, Benoit VANNIER [1]<benoit.vannier [at] apog>wrote
:

Hello,

Iperf is pretty good at this ... It s free

Ben

-----Message d'origine-----
De : Mark Urbach [[2]mailto:mark.urbach [at] pnpt]
Envoyé : lundi 2 novembre 2009 22:57
À : [3]nanog [at] nanog
Objet : Speed Testing and Throughput testing

Anyone have a good solution to get "accurate" speed results when testing at
10/100/1000 Ethernet speeds?

Do you have a server/software that customer can test too?

Thanks,
Mark Urbach
PinPoint Communications, Inc.
100 N. 12th St Suite 500
Lincoln, NE 68508
402-438-6211 ext 1923 Office
402-660-7982 Cell
[4]mark.urbach [at] pnpt
[[5]cid:image003.jpg [at] 01CA5BD5]

References

1. mailto:benoit.vannier [at] apog
2. mailto:mark.urbach [at] pnpt
3. mailto:nanog [at] nanog
4. mailto:mark.urbach [at] pnpt
5. cid:image003.jpg [at] 01CA5BD5

wmaton at ryouko

Nov 3, 2009, 4:19 AM

Post #13 of 20 (1120 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

On Tue, 3 Nov 2009, Jason Biel wrote:

> Please take note with using iperf that you'll want to make sure the
> appropriate TCP Window Size has been negotiated. We recently did some
> testing with systems that had decided to pick less than optimal window sizes
> and in turn had to manually set the size within iperf options.

Indeed this is true.

Also, if you use one of the Internet2 network test web100-enabled servers,
you can try testing through a web browser. There's both NPAD and NDT on
distributed on different nodes, although each has its own slightly
different tests. It's also not a bad set of tools for support people
wanting to troubleshoot bandwidth problems caused by duplex misconfigs.

>
> Jason
>
> On Tue, Nov 3, 2009 at 4:01 AM, Benoit VANNIER <benoit.vannier [at] apog>wrote:
>
>> Hello,
>>
>> Iperf is pretty good at this ... It s free
>>
>>
>> Ben
>>
>>
>> -----Message d'origine-----
>> De : Mark Urbach [mailto:mark.urbach [at] pnpt]
>> Envoyé : lundi 2 novembre 2009 22:57
>> À : nanog [at] nanog
>> Objet : Speed Testing and Throughput testing
>>
>> Anyone have a good solution to get "accurate" speed results when testing at
>> 10/100/1000 Ethernet speeds?
>>
>> Do you have a server/software that customer can test too?
>>
>>
>>
>> Thanks,
>> Mark Urbach
>> PinPoint Communications, Inc.
>> 100 N. 12th St Suite 500
>> Lincoln, NE 68508
>> 402-438-6211 ext 1923 Office
>> 402-660-7982 Cell
>> mark.urbach [at] pnpt
>> [cid:image003.jpg [at] 01CA5BD5]
>>
>>
>>
>
>
> --
> Jason Biel
>

wfms

jason at biel-tech

Nov 3, 2009, 4:49 AM

Post #14 of 20 (1115 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

Linux always worked best for us as well, was easy running a livecd with
laptops. We found that two windows XP machines, same identical hardware and
OS load yielded different registry settings (or lack thereof) for TCP Window
setting.

Jason

On Tue, Nov 3, 2009 at 6:18 AM, Bret Clark <bclark [at] spectraaccess> wrote:

> True, we usually find Linux based machines work better running IPerf
> then Windows (at least out of the box) because of the TCP window
> size....well Windows XP at least, don't know about Vista or 7.
> Jason Biel wrote:
>
> Please take note with using iperf that you'll want to make sure the
> appropriate TCP Window Size has been negotiated. We recently did some
> testing with systems that had decided to pick less than optimal window
> sizes
> and in turn had to manually set the size within iperf options.
>
> Jason
>
> On Tue, Nov 3, 2009 at 4:01 AM, Benoit VANNIER [1]<benoit.vannier [at] apog
> >wrote
> :
>
>
> Hello,
>
> Iperf is pretty good at this ... It s free
>
>
> Ben
>
>
> -----Message d'origine-----
> De : Mark Urbach [[2]mailto:mark.urbach [at] pnpt]
> Envoyé : lundi 2 novembre 2009 22:57
> À : [3]nanog [at] nanog
> Objet : Speed Testing and Throughput testing
>
> Anyone have a good solution to get "accurate" speed results when testing at
> 10/100/1000 Ethernet speeds?
>
> Do you have a server/software that customer can test too?
>
>
>
> Thanks,
> Mark Urbach
> PinPoint Communications, Inc.
> 100 N. 12th St Suite 500
> Lincoln, NE 68508
> 402-438-6211 ext 1923 Office
> 402-660-7982 Cell
> [4]mark.urbach [at] pnpt
> [[5]cid:image003.jpg [at] 01CA5BD5]
>
> References
>
> 1. mailto:benoit.vannier [at] apog
> 2. mailto:mark.urbach [at] pnpt
> 3. mailto:nanog [at] nanog
> 4. mailto:mark.urbach [at] pnpt
> 5. cid:image003.jpg [at] 01CA5BD5
>

--
Jason Biel

mhelmest at uvic

Nov 4, 2009, 2:49 PM

Post #15 of 20 (1094 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

We had a problem where our (mostly research network connected, international) users were getting generally low HTTP transfer speeds, even though the path was often gigabit. The classic high bandwidth/high latency problem.

Initially I tried using iperf/ndt and friends but found that iperf required too much user knowledge and interaction, and NDT was sometimes inaccurate at diagnosing problems -- it seemed to be overly fond of saying there was a duplex mismatch or congestion. Iperf in TCP mode either requires manually seeking the number of streams to try and find optimum throughput, or doing window size tweaks.

I also found that packet captures were useful for discovering problems in the path; you can load it up in wireshark or tcptrace and get a sequence no. vs time graph, look for packetloss, or other good things like that.

Anyways I didn't find much out there in terms of automating this type of thing (simple throughput tests with packet capture) so I just ended up making my own. It does a dump of 10 sec. of test traffic, uses a somewhat dumb algorithm to seek up the number of TCP streams, and gets an AS path from a BGP route server and displays it to the user. The caveat is that it only tests your download speed, not upload, since that was primarily what I was interested in.

You can give it a try at: http://caranthir.dao.nrc.ca/netperf-www/ (login nanog/nanog). User guide here: http://www.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/netperf/testdetail.shtml

I might end up packaging and releasing the code if there is interest.

On Mon, 2 Nov 2009 15:56:56 -0600
Mark Urbach <mark.urbach [at] pnpt> wrote:

> Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?
>
> Do you have a server/software that customer can test too?
>
>
>
> Thanks,
> Mark Urbach
> PinPoint Communications, Inc.
> 100 N. 12th St Suite 500
> Lincoln, NE 68508
> 402-438-6211 ext 1923 Office
> 402-660-7982 Cell
> mark.urbach [at] pnpt
> [cid:image003.jpg [at] 01CA5BD5]
>
>

horms at verge

Nov 4, 2009, 8:54 PM

Post #16 of 20 (1091 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

On Tue, Nov 03, 2009 at 06:49:05AM -0600, Jason Biel wrote:
> Linux always worked best for us as well, was easy running a livecd with
> laptops. We found that two windows XP machines, same identical hardware and
> OS load yielded different registry settings (or lack thereof) for TCP Window
> setting.

I would have thought that (netperf's) UDP tests were more appropriate
than TCP for testing link speed. Am I missing the point?

ml at axmo12

Nov 5, 2009, 2:30 AM

Post #17 of 20 (1087 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

Hi Michael,

Zitat von Michael Helmeste <mhelmest [at] uvic>:
> [...]
> You can give it a try at: http://caranthir.dao.nrc.ca/netperf-www/
> (login nanog/nanog). User guide here:
> http://www.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/netperf/testdetail.shtml
>
> I might end up packaging and releasing the code if there is interest.

If you have time it would be great if you could do that. - I'm very
interested in it!

Kind Regards,
Axel

jason at biel-tech

Nov 5, 2009, 6:07 AM

Post #18 of 20 (1085 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

We actually used both tests to finally narrow down the TCP window size
issue. We first used the UDP test to make sure the link was good.

Jason

On Wed, Nov 4, 2009 at 10:54 PM, Simon Horman <horms [at] verge> wrote:

> On Tue, Nov 03, 2009 at 06:49:05AM -0600, Jason Biel wrote:
> > Linux always worked best for us as well, was easy running a livecd with
> > laptops. We found that two windows XP machines, same identical hardware
> and
> > OS load yielded different registry settings (or lack thereof) for TCP
> Window
> > setting.
>
> I would have thought that (netperf's) UDP tests were more appropriate
> than TCP for testing link speed. Am I missing the point?
>
>

--
Jason Biel

jason at i6ix

Nov 5, 2009, 6:26 AM

Post #19 of 20 (1081 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

ml [at] axmo12 wrote:
> Hi Michael,
>
> Zitat von Michael Helmeste <mhelmest [at] uvic>:
>> [...]
>> You can give it a try at: http://caranthir.dao.nrc.ca/netperf-www/
>> (login nanog/nanog). User guide here:
>> http://www.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/netperf/testdetail.shtml
>>
>> I might end up packaging and releasing the code if there is interest.
>
> If you have time it would be great if you could do that. - I'm very
> interested in it!
+1

oberman at es

Nov 7, 2009, 3:56 PM

Post #20 of 20 (1017 views)
Permalink

Re: Speed Testing and Throughput testing [In reply to]

> Date: Mon, 2 Nov 2009 18:22:19 -0500
> From: Jon Meek <meekjt [at] gmail>
>
> I use iperf with packet capture on both sides, then analyze the packet
> capture for per-second throughput and re-transmits. I usually do 10
> TCP streams for 30 seconds.
>
> Note that on GigE with significant RTTs (5-15 ms) some TCP tuning is
> needed to deal with the bandwidth delay product. It is also possible
> that Ethernet drivers will have an effect. Local testing of the pair
> of test machines should be done if you can't get to about 980 Mbps on
> a Gig link (keeping in mind the comment about TCP tuning as latency
> increases).
>
> Jon
>
> On Mon, Nov 2, 2009 at 4:56 PM, Mark Urbach <mark.urbach [at] pnpt> wrote:
> > Anyone have a good solution to get "accurate" speed results when testing at 10/100/1000 Ethernet speeds?
> >
> > Do you have a server/software that customer can test too?

I'll also suggest http://fasterdata.es.net as a resource for network
tuning. Tuning TCP is hard. UDP is simple, but some things can even
impact UDP.

Many less than obvious things can have a huge impact on high-speed data
transfer. The choice of congestion algorithms can be very
significant. As anyone who has used bittorrent should have noticed,
having multiple TCP streams works better than a single stream.

An oddity we have noted is that some routers will process switch layer
2 traffic if a layer 3 interface exists on the port even if it is
unconfigured and unused. Man, that kills performance, even in low
latency situations!

FWIW, we use mostly iperf, but may be biased as the iperf maintainer
works here. We did start using iperf before we hired him, though.
--
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman [at] es Phone: +1 510 486-8634
Key fingerprint:059B 2DDF 031C 9BA3 14A4 EADA 927D EBB3 987B 3751

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads