<html><head><meta http-equiv="Content-Type" content="text/html charset=us-ascii"><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto" style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div>If you create a class inherited from ns3::ndn::fw::Nack, then you can overload OnNack event and do custom processing within it (don't forget to call parent's class processing as well). There are other events</div><div><br></div><div>Normally, you shouldn't need to modify anything inside ndnSIM and implement everything inside the extension.</div><div><br></div><div>There is a basic example on the website (<a href="http://ndnsim.net/fw.html#extending-strategy">http://ndnsim.net/fw.html#extending-strategy</a>) that shows what can be done in the strategy extension.<br><br>---<div>Alex</div></div><div><br>On Feb 7, 2013, at 11:23 AM, Yifeng Li <<a href="mailto:yifengl@email.arizona.edu">yifengl@email.arizona.edu</a>> wrote:<br><br></div><blockquote type="cite">Hello Alex,<br><br>If I were to incorporate NACKs in the forwarding scheme, e.g. receiving NACK => window reduce, where would I modify that? Would it still be within the forwarding strategy extension or should I modify the function that handles NACKs in general? If it's the latter, where is this function?<br><br>Thanks for your help,<br>Yifeng<br><br><br><div class="gmail_quote">On Thu, Feb 7, 2013 at 12:13 PM, Alex Afanasyev <span dir="ltr"><<a href="mailto:afanasev@cs.ucla.edu" target="_blank">afanasev@cs.ucla.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div>Hi Yifeng,</div><div>
<br></div><div>Actual sending is happening in the base ndn::Consumer class (ndn-consumer.cc). As this is an application, the only communication channel to/from l3 protocol is through application face and registered callback (actually, to send from app, you always need to use callback).</div>
<div><br></div><div>Wouldn't extension of the forwarding strategy be a better fit for your needs? Inside the forwarding strategy you have full control to which faces you want to send messages.</div><div><br></div><div>
--</div><div>Alex</div><div><div class="h5"><div><br></div><div><div><div>On Feb 7, 2013, at 11:08 AM, Yifeng Li <<a href="mailto:yifengl@email.arizona.edu" target="_blank">yifengl@email.arizona.edu</a>> wrote:</div>
<br><blockquote type="cite">Hi Alex,<br><br>I have a question regarding the ndnSIM apps. I'm trying simulate something similar to hop by hop TCP, e.g. every node receives interests, buffers them and then sends them out with typical TCP fashion (AIMD). I'm looking at ndn-consumer-window.cc, specifically in ConsumerWindow::ScheduleNextPacket. I see a send event is scheduled, and Consumer::SendPacket is called, but where in SendPacket is the packet actually sent? Can I specify a face to send? Also when receiving interests, are the interests buffered somewhere?<br>
<br>Thanks for your help,<br>Yifeng<br><br><br><br><div class="gmail_quote">On Wed, Nov 28, 2012 at 12:19 PM, Alex Afanasyev <span dir="ltr"><<a href="mailto:afanasev@cs.ucla.edu" target="_blank">afanasev@cs.ucla.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">Yes. Delay should not be too large, otherwise Reno is not able to get the right value for the congestion window (because it is increasing by one every RTT). Though 200ms is not that large number for RTT, especially for the 1Mbps links...<br>
<br>
---<br>
Alex<br>
<div><br>
On Nov 28, 2012, at 11:08 AM, Yifeng Li <<a href="mailto:yifengl@email.arizona.edu" target="_blank">yifengl@email.arizona.edu</a>> wrote:<br>
<br>
> Hello Alex,<br>
><br>
> Regarding my last email, if I change the delay on each link to 10ms (instead of approx 130ms), the throughput will look similar to the ones displayed in the forwarding paper.<br>
> Sincerely,<br>
> Yifeng<br>
><br>
><br>
><br>
><br>
> On Wed, Nov 28, 2012 at 11:25 AM, Yifeng Li <<a href="mailto:yifengl@email.arizona.edu" target="_blank">yifengl@email.arizona.edu</a>> wrote:<br>
> Hello Alex,<br>
><br>
> Thanks for the tutorial, I will look into it. As for the TCP part of the experiment, the graph that I got from using pcap seems a little different from what is displayed in the forward paper. It seems like the throughput never reaches 1Mbps after the initial back off. I've attached the throughput displayed via wireshark to this email. I assume the default TCP used by ns3 is TCP Reno. I'll also attach my experiment to this email as well.<br>
><br>
> Thanks for your help,<br>
> Yifeng<br>
><br>
<br>
</div></blockquote></div><br>
</blockquote></div><br></div></div></div></div></blockquote></div><br>
</blockquote></body></html>