I think I should make something a bit more clear regarding this post.  In
that I am working with a mesh network, I am running these tests over
multiple hops.  For the previous email I sent, I was transferring over 2
hops:

A->B->C

I doubt that I am at the capacity of the network card, because when I cut
the number of hops down to 1:

A->B

I am able to achieve a much high send rate.  Here is the same 10Mbps report,
only this time for one hop (and yes I am running this many times to make
sure this isn't an anomaly) :
------------------------------------------------------------
Client connecting to 5.0.0.9, UDP port 4000
Sending 1470 byte datagrams
UDP buffer size:   112 KByte (default)
------------------------------------------------------------
[  3] local 5.0.0.1 port 59795 connected with 5.0.0.9 port 4000
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-10.0 sec  12209 KBytes  10000 Kbits/sec
[  3] Sent 8505 datagrams
[  3] Server Report:
[  3]  0.0-10.0 sec  12198 KBytes  9991 Kbits/sec  0.327 ms    7/ 8504
(0.082%)
[  3]  0.0-10.0 sec  1 datagrams received out-of-order


It seems as though the longer my chain of nodes are, the lower my max send
rate can be.  The issue is that I am running UDP, which, unlike TCP, doesn't
care about reliability (and therefore doesn't send any ACKs).  So, why is
iperf (or something else) causing my maximum send rate to suffer?  I will be
posting the plateau of one hop as soon as the tests are done.

Randy



On Tue, Apr 27, 2010 at 10:04 AM, Randy Buck <sutekistud...@gmail.com>wrote:

> Hi,
>
> I'm working on a *wireless* mesh network and am trying to use iperf to
> benchmark performance.  I'm running into an issue that has been very
> troublesome to understand.  I am using the following setup:
>
> OS: Ubuntu 9.10
> Wireless chipset: Aetheros
> Wireless driver: ath5k (included in 9.10)
> Let me know if I've left anything out....
>
>
> Here is how I am starting the server:
> iperf -s -u -p 4000
>
> Here is what I am running on the client:
> iperf -u -fk -c 5.0.0.9 -b 10000K -p 4000 -t 10
>
> Note that I understand that the bandwidth is really 10 Mbps, I'm simply
> leaving in terms of Kbps for graphing purposes.
>
> Here is my output:
>
> ------------------------------------------------------------
> Client connecting to 5.0.0.9, UDP port 4000
> Sending 1470 byte datagrams
> UDP buffer size:   112 KByte (default)
> ------------------------------------------------------------
> [  3] local 5.0.0.5 port 38669 connected with 5.0.0.9 port 4000
> [ ID] Interval       Transfer     Bandwidth
> [  3]  0.0-10.0 sec  8343 KBytes  *6834 Kbits/sec*
> [  3] Sent 5812 datagrams
> [  3] Server Report:
> [  3]  0.0-10.1 sec  8137 KBytes  6631 Kbits/sec  0.698 ms  143/ 5811
> (2.5%)
> [  3]  0.0-10.1 sec  1 datagrams received out-of-order
>
> My concern is that I am not sending at 10 Mbps, but rather at 6.8 Mbps.  I
> need to be able to send at 10 Mbp; I don't care what the packet loss is, I
> just need to be able to see what it is in terms of what I configured iperf
> to send at (10 Mbps).
>
> When I run over the *wired* network, I get what I expect (namely that I
> always send at the correct sending rate).  The send rate is much higher for
> the wired network:
>
> Here is how I am starting the server:
> iperf -s -u -p 4000
>
> Here is what I am running on the client:
> iperf -u -fk -c mesh9 -b 200000K -p 4000 -t 10
> ------------------------------------------------------------
> Client connecting to mesh9, UDP port 4000
> Sending 1470 byte datagrams
> UDP buffer size:   112 KByte (default)
> ------------------------------------------------------------
> [  3] local 192.168.21.205 port 56030 connected with 192.168.21.209 port
> 4000
> [ ID] Interval       Transfer     Bandwidth
> [  3]  0.0-10.0 sec  247418 KBytes  *202684 Kbits/sec*
> [  3] Sent 172351 datagrams
> [  3] Server Report:
> [  3]  0.0-10.3 sec  117083 KBytes  93574 Kbits/sec  13.143 ms 90780/172340
> (53%)
> [  3]  0.0-10.3 sec  1 datagrams received out-of-order
>
>
> Why does iperf (or my wireless card, or my OS, or something else) not send
> at the full rate when transferring wirelessly?  I'm beginning to believe
> that iperf is not the tool to use for wireless measurements.  Are there any
> better solutions?
> Thanks for all your help,
>
> Randy
>
>
------------------------------------------------------------------------------
_______________________________________________
Iperf-users mailing list
Iperf-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/iperf-users

Reply via email to