<br><br><div><span class="gmail_quote">On 7/6/07, <b class="gmail_sendername">V Anil Kumar</b> <<a href="mailto:anil@cmmacs.ernet.in">anil@cmmacs.ernet.in</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>Hi,<br><br>What are the buffer sizes configured at the interface level on<br>intermediate routers? Are they also set to the BDP of the link? Also, what<br>is the buffer management scheme (RED or FIFO) on the intermediate routers?
<br></blockquote><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">These parameters will also have an impact on the end-to-end throughput.</blockquote>
<div><br style="color: rgb(204, 102, 0);"><span style="color: rgb(204, 102, 0);"> The network is not under my administration . So , I will</span><br style="color: rgb(204, 102, 0);"><span style="color: rgb(204, 102, 0);">
probably ask my metwork provider to give me the result</span>. <br> <br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">The configuration you attached shows that you could improve the throughput
<br>from 55 mbps to 95 mbps by decreasing the buffer size from 921600 to<br>175636. While your text says that you achieved 95 mbps by increasing the<br>buffer size.</blockquote><div><br> <span style="color: rgb(204, 102, 0);">
Sorry, that was a typo . It should have been 1756366. (nearly 1.7 MB ). </span><br><br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Any idea why a dedicated 100 mbps link gives an RTT of 70 msec. under no<br>load condition?<br><br> <span style="color: rgb(255, 102, 0);"> The two end points are very far away. It is more than 3000 kms . So, probably</span>
</blockquote><div><span style="color: rgb(255, 102, 0);"> that might me the reason. The RTT is obeying the SLA parameters that was</span><br style="color: rgb(255, 102, 0);"><span style="color: rgb(255, 102, 0);"> defined for this network.
</span><br style="color: rgb(255, 102, 0);"><br><br>With Thanks and Regards<br> zaman<br> <br></div><br><div><br> <br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Anil<br><br> On Thu, 5 Jul 2007, query wrote:<br><br>> Hi All,<br>><br>> I was doing some Bandwidth measurement test on a 100 mbs link with a RTT<br>> of about 70ms.<br>> Based on that, I calculated the BDP as follows.
<br>><br>> BDP = Bandwidth * RTT<br>> = 13107200 bytes * .07 secs<br>> = 896 Kbytes<br>> = 900 Kbyes (approx)<br>> = 921600 bytes<br>>
<br>><br>> After that I adjusted the TCP window size as follows:<br>><br>> /proc/sys/net/core/rmem_max 921600<br>> /proc/sys/net/core/wmem_max 921600<br>> /proc/sys/net/ipv4/tcp_rmem 4096 87380 921600
<br>> /proc/sys/net/ipv4/tcp_wmem 4096 87380 921600<br>><br>> These adjustmenst I had done on a Linux host with 2.6.15 kernel. The<br>> congestion<br>> control algorithm , it is using BIC
<br>><br>> The same window adjustments I had performed on the other hand on a Linux<br>> host<br>> with kernel 2.6.9 . It is also using BIC for congestion control.<br>><br>> The Bandwidth Performnce test I am doing using iperf , a highly popular
<br>> public<br>> domain tool for measuring TCP & UDP Bandwidth Performance .<br>><br>> With the default Linux 2.6 TCP window settings , I was getting a<br>> throughput of<br>> nearly 10mbs which is very low for a 100 mbs link.
<br>><br>> So , I performed the above TCP adjustmets and I found the throughput to<br>> be around<br>> 55 mbs which is a significant improvement . But that is not fully<br>> utilsing the link<br>> as it is a dedicated link and there was no other traffic.
<br>> This , I proved with the next experiment where I reached a link<br>> utilisation of<br>> little more than 95 mbs . That is very much O.K for a 100mbs link.<br>><br>> I did the following adjustments. I increased the above calculated BDP by
<br>> nearly<br>> half of the value . The TCP settings now look like this.<br>><br>> /proc/sys/net/core/rmem_max 175636<br>> /proc/sys/net/core/wmem_max 175636<br>> /proc/sys/net/ipv4/tcp_rmem 4096 87380 175636
<br>> /proc/sys/net/ipv4/tcp_wmem 4096 87380 175636<br>><br>> After these settings , I find the link utilisation to be nearly 95 mbs.<br>><br>> According to many papers that I read , I found that the BDP should be
<br>> equal<br>> to the product of Bandwidth * RTT .<br>> I had done that , but the link utilisation is only 50%. But when I<br>> increased<br>> to a much higher value , the link utilisation is nearly around 95 %.
<br>><br>> I am confused regarding my findings.Please clarify me so that I can<br>> perform<br>> the experiment correctly.<br>><br>> With Thanks in Advance<br>><br><br><br></blockquote></div>
<br>