More results: Linux iperf client, Windows iperf server:<div><div># iperf -i 5 -t 15 -c 192.168.27.23 -u -b 700M</div><div>------------------------------------------------------------</div><div>Client connecting to 192.168.27.23, UDP port 5001</div>

<div>Sending 1470 byte datagrams</div><div>UDP buffer size:   126 KByte (default)</div><div>------------------------------------------------------------</div><div>[  3] local 192.168.27.10 port 56210 connected with 192.168.27.23 port 5001</div>

<div>[ ID] Interval       Transfer     Bandwidth</div><div>[  3]  0.0- 5.0 sec    428 MBytes    718 Mbits/sec</div><div>[  3]  5.0-10.0 sec    408 MBytes    684 Mbits/sec</div><div>[  3]  0.0-15.0 sec  1.22 GBytes    700 Mbits/sec</div>

<div>[  3] Sent 892738 datagrams</div><div>[  3] Server Report:</div><div>[  3]  0.0-15.0 sec  1.16 GBytes    663 Mbits/sec  0.015 ms 47569/892392 (5.3%)</div><div>[  3]  0.0-15.0 sec  1 datagrams received out-of-order</div>

<div><br></div><div>Reversed:</div><div><div>C:\Temp&gt;iperf -u -t 15 -c 192.168.27.10 -b 750M</div><div>------------------------------------------------------------</div><div>Client connecting to 192.168.27.10, UDP port 5001</div>

<div>Sending 1470 byte datagrams</div><div>UDP buffer size: 8.00 KByte (default)</div><div>------------------------------------------------------------</div><div>[1912] local 192.168.27.23 port 2100 connected with 192.168.27.10 port 5001</div>

<div>[ ID] Interval       Transfer     Bandwidth</div><div>[1912]  0.0-15.0 sec   174 MBytes  97.5 Mbits/sec</div><div>[1912] Server Report:</div><div>[1912]  0.0-15.0 sec   174 MBytes  97.5 Mbits/sec  1.055 ms    0/124429 (0%)</div>

<div>[1912] Sent 124429 datagrams</div><div><br>And then, running the iperf client on the same hardware as was running Windows, but running Ubuntu 9.10 (dual boot):<br>$ iperf -t 15 -c 192.168.27.10 -u -b 750M<br>------------------------------------------------------------<br>
Client connecting to 192.168.27.10, UDP port 5001<br>Sending 1470 byte datagrams<br>UDP buffer size:   112 KByte (default)<br>------------------------------------------------------------<br>[  3] local 192.168.27.23 port 51363 connected with 192.168.27.10 port 5001<br>
[ ID] Interval       Transfer     Bandwidth<br>[  3]  0.0-15.0 sec    316 MBytes    177 Mbits/sec<br>[  3] Sent 225456 datagrams<br>[  3] Server Report:<br>[  3]  0.0-15.0 sec    316 MBytes    177 Mbits/sec  0.207 ms    0/225455 (0%)<br>
[  3]  0.0-15.0 sec  1 datagrams received out-of-order<br><br>And with TCP:<br>$ iperf -t 15 -c 192.168.27.10<br>------------------------------------------------------------<br>Client connecting to 192.168.27.10, TCP port 5001<br>
TCP window size: 16.0 KByte (default)<br>------------------------------------------------------------<br>[  3] local 192.168.27.23 port 53251 connected with 192.168.27.10 port 5001<br>[ ID] Interval       Transfer     Bandwidth<br>
[  3]  0.0-15.0 sec    368 MBytes    206 Mbits/sec<br><br><br>The source hardware seems to be having a problem sending. Receiving, less so:<br>$ iperf -s<br>------------------------------------------------------------<br>
Server listening on TCP port 5001<br>TCP window size: 85.3 KByte (default)<br>------------------------------------------------------------<br>[  4] local 192.168.27.23 port 5001 connected with 192.168.27.10 port 37472<br>
[ ID] Interval       Transfer     Bandwidth<br>[  4]  0.0-15.0 sec    949 MBytes    531 Mbits/sec<br><br>$ iperf -t 15 -c 192.168.27.23 -u -b 750M<br>------------------------------------------------------------<br>Client connecting to 192.168.27.23, UDP port 5001<br>
Sending 1470 byte datagrams<br>UDP buffer size:   126 KByte (default)<br>------------------------------------------------------------<br>[  3] local 192.168.27.10 port 37815 connected with 192.168.27.23 port 5001<br>[ ID] Interval       Transfer     Bandwidth<br>
[  3]  0.0-15.0 sec  1.37 GBytes    783 Mbits/sec<br>[  3] Sent 999330 datagrams<br>[  3] Server Report:<br>[  3]  0.0-15.2 sec  1.09 GBytes    613 Mbits/sec  15.609 ms 204247/999327 (20%)<br>[  3]  0.0-15.2 sec  1 datagrams received out-of-order<br>
<br>...Although push the bandwidth (UDP) too high, and a lot of packets get lost.<br><br>Any further thoughts?<br><br>Kevin<br></div></div><br><div class="gmail_quote">On Mon, Apr 5, 2010 at 9:52 PM, Sean Walberg <span dir="ltr">&lt;<a href="mailto:swalberg@gmail.com" target="_blank">swalberg@gmail.com</a>&gt;</span> wrote:<br>

<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">IIRC UDP needs you to pass the desired bandwidth, otherwise it defaults to a megabit.<br><font color="#888888"><br>
Sean</font><div>
<div></div><div><br><br><div class="gmail_quote">On Mon, Apr 5, 2010 at 9:43 PM, Kevin McGregor <span dir="ltr">&lt;<a href="mailto:kevin.a.mcgregor@gmail.com" target="_blank">kevin.a.mcgregor@gmail.com</a>&gt;</span> wrote:<br>


<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">When I choose UDP, I get<div> 0.0-15.0 sec  1.88 MBytes  1.05 Mbits/sec  6.154 ms    0/ 1339 (0%)</div>


<div><br></div><div>No packet loss. 0.1% utilization!<div><div></div><div><br><br><div class="gmail_quote">On Mon, Apr 5, 2010 at 8:12 PM, Sean Walberg <span dir="ltr">&lt;<a href="mailto:swalberg@gmail.com" target="_blank">swalberg@gmail.com</a>&gt;</span> wrote:<br>



<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">On a LAN, window sizes aren&#39;t going to make that much of a difference.<div><br></div><div>

First look for the usual suspects - errors on the swich port or NIC.</div>
<div><br></div><div>If you can grab a copy of the traffic, such as with tcpdump -w test.pcap tcp port 5001, you can pull it into wireshark and look for TCP zero window conditions, retransmits, and duplicate ACKs. The TCP performance graph will also show whether or not the transmission is stalling.</div>




<div><br></div><div>Is the performance similar using the UDP test? Does it show loss?</div><div><br></div><div>Sean<br><br><div class="gmail_quote"><div><div></div><div>On Mon, Apr 5, 2010 at 7:27 PM, Kevin McGregor <span dir="ltr">&lt;<a href="mailto:kevin.a.mcgregor@gmail.com" target="_blank">kevin.a.mcgregor@gmail.com</a>&gt;</span> wrote:<br>




</div></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div><div></div><div>I&#39;ve been trying to narrow down where my performance bottleneck(s) is (are). I just ran &quot;iperf&quot; from a Windows PC (XP SP3) to my Ubuntu 9.10 server. With iperf in &quot;server&quot; mode on the PC, and the client running on the Ubuntu machine, I get<div>





 0.0-60.0 sec  4.23 GBytes   606 Mbits/sec</div><div><br></div><div>Reversed (iperf &quot;server&quot; running on Ubuntu server, client running on the PC), I get</div><div> 0.0-60.0 sec  1.13 GBytes   162 Mbits/sec</div>




<div>
<br></div><div>Wha...?? Both client and server runs on the PC report TCP window size to be 8 KB, but the Linux client reports 22.4 KB and the Linux server reports 85.3 KB. Increasing that to 256K on both ends has little effect. Does anyone have any suggestions?</div>





<div><br></div><div>The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?</div>





<div><br></div><font color="#888888"><div>Kevin</div>
</font><br></div></div>_______________________________________________<br>
Roundtable mailing list<br>
<a href="mailto:Roundtable@muug.mb.ca" target="_blank">Roundtable@muug.mb.ca</a><br>
<a href="http://www.muug.mb.ca/mailman/listinfo/roundtable" target="_blank">http://www.muug.mb.ca/mailman/listinfo/roundtable</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br>Sean Walberg &lt;<a href="mailto:sean@ertw.com" target="_blank">sean@ertw.com</a>&gt;    <a href="http://ertw.com/" target="_blank">http://ertw.com/</a><br>
</div>
<br>_______________________________________________<br>
Roundtable mailing list<br>
<a href="mailto:Roundtable@muug.mb.ca" target="_blank">Roundtable@muug.mb.ca</a><br>
<a href="http://www.muug.mb.ca/mailman/listinfo/roundtable" target="_blank">http://www.muug.mb.ca/mailman/listinfo/roundtable</a><br>
<br></blockquote></div><br></div></div></div>
<br>_______________________________________________<br>
Roundtable mailing list<br>
<a href="mailto:Roundtable@muug.mb.ca" target="_blank">Roundtable@muug.mb.ca</a><br>
<a href="http://www.muug.mb.ca/mailman/listinfo/roundtable" target="_blank">http://www.muug.mb.ca/mailman/listinfo/roundtable</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br>Sean Walberg &lt;<a href="mailto:sean@ertw.com" target="_blank">sean@ertw.com</a>&gt;    <a href="http://ertw.com/" target="_blank">http://ertw.com/</a><br>
</div></div><br>_______________________________________________<br>
Roundtable mailing list<br>
<a href="mailto:Roundtable@muug.mb.ca" target="_blank">Roundtable@muug.mb.ca</a><br>
<a href="http://www.muug.mb.ca/mailman/listinfo/roundtable" target="_blank">http://www.muug.mb.ca/mailman/listinfo/roundtable</a><br>
<br></blockquote></div><br></div>