<div dir="ltr">Hi Peter<div><br></div><div>With 30 Interests/second, NFD2 is unlikely to be overloaded.</div><div><br></div><div>What's the MTU used on both links?</div><div>If it's below 9000, NDNLP fragmentation or IP fragmentation may cause delays.</div><div>Please set MTU to 9000 or above and try again.</div><div><br></div><div>Yours, Junxiao</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jan 22, 2015 at 2:04 PM, Gusev, Peter <span dir="ltr"><<a href="mailto:peter@remap.ucla.edu" target="_blank">peter@remap.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">
Hi Junxiao,
<div><br>
</div>
<div>Thanks for your input!</div>
<div>I'm surprised to know that NFD2 was overloaded - this hub did not run any extraneous apps except MacOS Activity Monitor and NFD itself, however NFD1 and NFD3 were executing NDN-RTC code (consumer and producer respectively).</div>
<div>Besides that, NDN-RTC was running on lowest configuration (having one 8000-bytes segment per frame) so the number of interests was minimal (~30 interests/sec), which means for higher quality, the number of interests will increase.</div>
<div><br>
</div>
<div>Thanks,</div>
<div><br>
<div><span class="">-- <br>
Peter Gusev<br>
<a href="mailto:peter@remap.ucla.edu" target="_blank">peter@remap.ucla.edu</a><br>
<a href="tel:%2B1%20213%205872748" value="+12135872748" target="_blank">+1 213 5872748</a> (USA)<br></span>
peetonn_ (skype) </div><div><div class="h5">
<br>
<div>
<blockquote type="cite">
<div>On Jan 21, 2015, at 11:03 PM, Burke, Jeff <<a href="mailto:jburke@remap.ucla.edu" target="_blank">jburke@remap.ucla.edu</a>> wrote:</div>
<br>
<div>
<div style="word-wrap:break-word;font-size:14px;font-family:Calibri,sans-serif">
<div>Hi Junxiao,</div>
<div>Thanks for looking at this. </div>
<div>Would this <50% CPU utilization target to apply to end-hosts as well? (Seems hard to meet this while running media apps.)</div>
<div>Jeff</div>
<span>
<div style="font-family:Calibri;font-size:11pt;text-align:left;border-width:1pt medium medium;border-style:solid none none;padding:3pt 0in 0in;border-top-color:rgb(181,196,223)">
<span style="font-weight:bold">From: </span>Junxiao Shi <<a href="mailto:shijunxiao@email.arizona.edu" target="_blank">shijunxiao@email.arizona.edu</a>><br>
<span style="font-weight:bold">Date: </span>Wed, 21 Jan 2015 23:39:03 -0700<br>
<span style="font-weight:bold">To: </span>"Gusev, Peter" <<a href="mailto:peter@remap.UCLA.EDU" target="_blank">peter@remap.UCLA.EDU</a>><br>
<span style="font-weight:bold">Cc: </span>"<<a href="mailto:nfd-dev@lists.cs.ucla.edu" target="_blank">nfd-dev@lists.cs.ucla.edu</a>>" <<a href="mailto:nfd-dev@lists.cs.ucla.edu" target="_blank">nfd-dev@lists.cs.ucla.edu</a>><br>
<span style="font-weight:bold">Subject: </span>Re: [Nfd-dev] NDN-RTC: NFD processing logs<br>
</div>
<div><br>
</div>
<blockquote style="BORDER-LEFT:#b5c4df 5 solid;PADDING:0 0 0 5;MARGIN:0 0 0 5">
<div dir="ltr">Hi Peter
<div><br>
</div>
<div>It appears that NFD2 node is overloaded.</div>
<div>There is a big delay between NFD3 returns first Data and NFD2 receives that Data.</div>
<div>There is a big delay between NFD1 forwards second Interests and NFD receives that Interest.</div>
<div><br>
</div>
<div>To determine whether a node is overloaded, look at its CPU utilization. If CPU utilization of any core is over 50%, reduce traffic rate.</div>
<div><br>
</div>
<div><br>
</div>
<div>It's expected for NFD2 not to return the second Data, because PIT entry has been satisfied by the first Data.</div>
<div>When the second Data arrives, PIT entry contains no downstream record, and therefore Data will not be returned to NFD1.</div>
<div><br>
</div>
<div><br>
</div>
<div>This experiment appears to be using an outdated NFD version (older than Sep 08, 2014).</div>
<div>For future reports, please use git-HEAD.</div>
<div><br>
</div>
<div>Yours, Junxiao<br>
<div class="gmail_extra"><br>
<div class="gmail_quote">On Wed, Jan 21, 2015 at 1:32 PM, Gusev, Peter <span dir="ltr">
<<a href="mailto:peter@remap.ucla.edu" target="_blank">peter@remap.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>
<div style="word-wrap:break-word">Hi all,
<div><br>
</div>
<div>I’ve been testing <a href="https://github.com/remap/ndnrtc/tree/issues/2283-retransmissions" target="_blank"></a><a href="https://github.com/remap/ndnrtc/tree/issues/2283-retransmissions" target="_blank">N</a>DN-RTC with new
consumer algorithm and exploring the problem of rebufferings on consumer side (a state, when consumer is not getting data for some amount of time, flushes everything and starts over again). Consumer rebufferings are perceived as audio/video interruptions for
1-1.5sec by the user and decrease overall user experience.</div>
<div><br>
</div>
<div>I'll appreciate your insights on the analysis of the logs I retrieved from NFD.</div>
<div><br>
</div>
<div><u>Setup:</u></div>
<div><br>
</div>
<div>[Consumer] <—> [<b>NFD1</b>] <==<i>WiFi_link</i>==> [<b>NFD2</b>] <==<i>Ethernet_link</i>==> [<b>NFD3</b>] <—> [Producer]</div>
<div><br>
</div>
<div><u>Case:</u></div>
<div>According to my logs, consumer starved for data after 2656 segment has arrived, so I tracked interests and data for
<b>2657</b> segment. According to my logs, consumer eventually received this data but too late and rebuffering was already inevitable.</div>
<div><br>
</div>
<div><b>NFD1 log:</b></div>
<div>
<div><u><i>initial interest:</i><span style="white-space:pre-wrap"></span></u><span style="white-space:pre-wrap"></span>1421288806.271931 DEBUG: [Forwarder] onIncomingInterest face=266 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.272215 DEBUG: [Forwarder] onOutgoingInterest face=262 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.272469 DEBUG: [BestRouteStrategy2] /ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00?ndn.MustBeFresh=1&ndn.InterestLifetime=5000&ndn.Nonce=108889115 from=266
newPitEntry-to=262</div>
<div><b><span style="white-space:pre-wrap"></span>^ 270ms</b></div>
<div><i><u>retransmission:</u></i><span style="white-space:pre-wrap"></span>1421288806.542289 DEBUG: [Forwarder] onIncomingInterest face=266 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.542697 DEBUG: [Forwarder] onOutgoingInterest face=262 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.543032 DEBUG: [BestRouteStrategy2] /ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00?ndn.MustBeFresh=1&ndn.InterestLifetime=5000&ndn.Nonce=562292066 from=266
retransmit-retry-to=262</div>
<div><span style="white-space:pre-wrap"></span><b>^ 286ms</b></div>
<div><i><u>data arrives:</u></i><span style="white-space:pre-wrap"><i><u></u></i></span>1421288806.829462 DEBUG: [Forwarder] onIncomingData face=262 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.829818 DEBUG: [Forwarder] onIncomingData matching=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.830042 DEBUG: [Strategy] beforeSatisfyPendingInterest pitEntry=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00 inFace=262 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.830449 DEBUG: [Forwarder] onOutgoingData face=266 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div style="font-weight:bold"><br>
</div>
<div><b>NFD2 log:</b></div>
<div>
<div><i><u>initial interest:</u></i><span style="white-space:pre-wrap"><u></u></span>1421288806.015075 DEBUG: [Forwarder] onIncomingInterest face=266 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.015845 DEBUG: [Forwarder] onOutgoingInterest face=264 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.016255 DEBUG: [BestRouteStrategy2] /ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00?ndn.MustBeFresh=1&ndn.InterestLifetime=5000&ndn.Nonce=108889115 from=266
newPitEntry-to=264</div>
<div><span style="white-space:pre-wrap"></span><b>^ 545ms</b></div>
<div><i><u>retransmission:</u></i><span style="white-space:pre-wrap"></span>1421288806.561516 DEBUG: [Forwarder] onIncomingInterest face=266 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.562233 DEBUG: [Forwarder] onOutgoingInterest face=264 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.562626 DEBUG: [BestRouteStrategy2] /ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00?ndn.MustBeFresh=1&ndn.InterestLifetime=5000&ndn.Nonce=562292066 from=266
retransmit-retry-to=264</div>
<div><span style="white-space:pre-wrap"></span><b>^ 2ms</b></div>
<div><i><u>data arrives:</u></i><span style="white-space:pre-wrap"></span>1421288806.564828 DEBUG: [Forwarder] onIncomingData face=264 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.565282 DEBUG: [Forwarder] onIncomingData matching=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.565594 DEBUG: [Strategy] beforeSatisfyPendingInterest pitEntry=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00 inFace=264 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.566213 DEBUG: [Forwarder] onOutgoingData face=266 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span><b>^ 13ms</b></div>
<div><i><u>data arrives 2:*</u></i><span style="white-space:pre-wrap"></span>1421288806.579547 DEBUG: [Forwarder] onIncomingData face=264 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.579915 DEBUG: [Forwarder] onIncomingData matching=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.580162 DEBUG: [Strategy] beforeSatisfyPendingInterest pitEntry=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00 inFace=264 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
</div>
<div><br>
</div>
<div><b>*NOTE:</b> <i>no data was forwarded</i></div>
<div><br>
</div>
<div><br>
</div>
<div><b>NFD3 log:</b></div>
<div>
<div><u><i>initial interest:</i></u><span style="white-space:pre-wrap"></span>1421288806.249712 DEBUG: [Forwarder] onIncomingInterest face=266 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.250177 DEBUG: [Forwarder] onOutgoingInterest face=269 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.250573 DEBUG: [BestRouteStrategy2] /ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00?ndn.MustBeFresh=1&ndn.InterestLifetime=5000&ndn.Nonce=108889115 from=266
newPitEntry-to=269</div>
<div><span style="white-space:pre-wrap"></span><b>^271ms</b></div>
<div><u><i>data generated:</i></u><span style="white-space:pre-wrap"></span>1421288806.521982 DEBUG: [Forwarder] onIncomingData face=269 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.522531 DEBUG: [Forwarder] onIncomingData matching=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.522844 DEBUG: [Strategy] beforeSatisfyPendingInterest pitEntry=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00 inFace=269 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span>1421288806.523363 DEBUG: [Forwarder] onOutgoingData face=266 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
<div><span style="white-space:pre-wrap"></span><b>^273ms</b></div>
<div><u><i>retransmission:</i></u><span style="white-space:pre-wrap"></span>1421288806.796022 DEBUG: [Forwarder] onIncomingInterest face=266 interest=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00</div>
<div><span style="white-space:pre-wrap"></span>1421288806.796418 DEBUG: [Forwarder] onOutgoingData face=266 data=/ndn/edu/ucla/remap/ndnrtc/user/remap/streams/camera/low/delta/2657/data/%00%00/1/2749/91/0</div>
</div>
<div><br>
</div>
<div><br>
</div>
<div><br>
</div>
<div>There are several questions arise:</div>
<div>1. Why do you think the delay b/w initial interest and retransmission for the NFD2 became 545ms (compared to 270ms with NFD1)? Could it be WiFi fault?</div>
<div>2. How come the delay b/w initial interest and data in NFD3 is 271ms and NFD3 forwards this data to NFD2, but NFD2 didn’t get this data and instead got retransmission interest after another 545-271=<b>274ms
</b>and later got the same data with 13ms difference? Could it be unreliable logging in NFD?</div>
<div><br>
</div>
<div>I attached full NFD logs to this message.</div>
<div><br>
</div>
<div>Thank you in advance.</div>
<div>All ideas and thoughts are welcome. </div>
<div><br>
</div>
<div>Thanks,</div>
<div><br>
</div>
<div>PS. Basically, such things block deployment of NDN-RTC (NdnCon) for the community as it delivers low-quality user experience (i.e. videoconference with interruptions =( ), so I really hope for the help and feedback.</div>
<span><font color="#888888">
<div><br>
</div>
<div></div>
</font></span></div>
<span><font color="#888888"></font></span></div>
<span><font color="#888888">
<div style="word-wrap:break-word">
<div>
<div></div>
<div><br>
<div>-- <br>
Peter Gusev<br>
<a href="mailto:peter@remap.ucla.edu" target="_blank">peter@remap.ucla.edu</a><br>
<a href="tel:%2B1%20213%205872748" value="+12135872748" target="_blank">+1 213 5872748</a> (USA)<br>
<a href="tel:%2B7%20916%204434826" value="+79164434826" target="_blank">+7 916 4434826</a> (Russia)<br>
<a href="tel:%2B37%20259%20226448" value="+37259226448" target="_blank">+37 259 226448</a> (in case any other number is unavailable)<br>
peetonn_ (skype)</div>
</div>
</div>
</div>
</font></span></div>
</blockquote>
</div>
</div>
</div>
</div>
_______________________________________________ Nfd-dev mailing list <a href="mailto:Nfd-dev@lists.cs.ucla.edu" target="_blank">
Nfd-dev@lists.cs.ucla.edu</a> <a href="http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev" target="_blank">
http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev</a> </blockquote>
</span></div>
</div>
</blockquote>
</div>
<br>
</div></div></div>
</div>
</blockquote></div><br></div>