[ndnSIM] Seeking help regarding ndnsim; intreset with different sequence number

Lixia Zhang lixia at CS.UCLA.EDU
Sun Apr 24 17:45:52 PDT 2016


> On Apr 24, 2016, at 1:04 PM, Ravi Kumar Singh <rk8000 at g.rit.edu> wrote:
> 
> Hi Lixia / Team,
> 
> I would like to create a custom cache replacement policy. I went through the LruPolicy class and I can see that it contains methods like evictEntries, insertToQueue, etc. However, it is not clear, when, how these methods are called. 
> 
> I believe in order to achieve what I want to, I would have to create a similar class. Am I on the right path here? Some starting point documentation of how the flow is would be really helpful.

I must admit that I dont know the code details. So we need Alex or Spyros to help out here.

> On Wed, Apr 20, 2016 at 3:14 AM, Ravi Kumar Singh <rk8000 at g.rit.edu <mailto:rk8000 at g.rit.edu>> wrote:
> Thanks Lixia/Spyridon! I will implement this and will let you the results! Thank you so much again!
> 
> On Wed, Apr 20, 2016 at 1:03 AM, Lixia Zhang <lixia at cs.ucla.edu <mailto:lixia at cs.ucla.edu>> wrote:
> 
>> On Apr 18, 2016, at 10:40 AM, Spyridon (Spyros) Mastorakis <mastorakis at cs.ucla.edu <mailto:mastorakis at cs.ucla.edu>> wrote:
>> 
>> Hi Ravi,
>> 
>> that looks right to me.
>> 
>> There is also something else that you need to do:
>> 
>> your consumer application should express Interests with the “mustBeFresh” selector set in order to retrieve only “fresh” data:
>> 
>> https://github.com/named-data-ndnSIM/ndn-cxx/blob/ndn-cxx-0.3.4-ndnSIM/src/interest.hpp#L417-L423 <https://github.com/named-data-ndnSIM/ndn-cxx/blob/ndn-cxx-0.3.4-ndnSIM/src/interest.hpp#L417-L423>
>> 
>> To this end you should do something like that:
>> 
>> Interest i(name);
>> i.setMustBeFresh(true);
>> ...
>>>> m_face->onReceiveInterest(*interest);
>> 
>> Hope that this helps.
>> 
>> Spyridon (Spyros) Mastorakis
>> Personal Website: http://cs.ucla.edu/~mastorakis/ <http://cs.ucla.edu/~mastorakis/>
>> Internet Research Laboratory
>> Computer Science Department
>> UCLA
> 
> copied the list here.
> Ravi, hopefully your issue was resolved?
> 
> 
>>> On Apr 17, 2016, at 11:54 PM, Ravi Kumar Singh <rk8000 at g.rit.edu <mailto:rk8000 at g.rit.edu>> wrote:
>>> 
>>> Hi Lixia,
>>> 
>>> No, I am not aware where we need to set this "must be fresh". I just followed the below link:
>>> 
>>> https://github.com/cawka/ndnSIM-examples/blob/master/examples/ndn-simple-with-content-freshness.cc <https://github.com/cawka/ndnSIM-examples/blob/master/examples/ndn-simple-with-content-freshness.cc>
>>> 
>>> and changed like below:
>>> 
>>> ccnxHelper.SetContentStore ("ns3::ndn::cs::Freshness::Lru","MaxSize", "2");
>>> 
>>> and 
>>> producerHelper.SetAttribute ("Freshness", TimeValue (Seconds (2.0)));
>>> 
>>> Please let me know where exactly we have to set "must be fresh" or anything we need to do?
>>> 
>>> Thanks and Regards,
>>> Ravi Kumar Singh
>>> 
>>> On Sun, Apr 17, 2016 at 11:25 PM, Lixia Zhang <lixia at cs.ucla.edu <mailto:lixia at cs.ucla.edu>> wrote:
>>> 
>>>> On Apr 17, 2016, at 12:27 PM, Ravi Kumar Singh <rk8000 at g.rit.edu <mailto:rk8000 at g.rit.edu>> wrote:
>>>> 
>>>> Hi Alex/Lixia/Team,
>>>> 
>>>> Thank you so much for the help! Now, I am trying to use the ns3::ndn::cs::Freshness::Lru   feature to use the freshness. But it seems its not giving desired results. Even after the freshness has expired the next interest is being served from the Cache server and not from the Origin Server. And as per the policy(ns3::ndn::cs::Freshness::Lru),it cache Data packets only for the time indicated by FreshnessPeriod.
>>> 
>>> did you set "must be fresh" in your interests packets?
>>> 
>>>> Thanks and Regards,
>>>> Ravi Singh
>>>> 
>>>> On Wed, Apr 13, 2016 at 9:41 PM, Alex Afanasyev <aa at cs.ucla.edu <mailto:aa at cs.ucla.edu>> wrote:
>>>> 
>>>>> On Apr 12, 2016, at 11:13 PM, Ravi Kumar Singh <rk8000 at g.rit.edu <mailto:rk8000 at g.rit.edu>> wrote:
>>>>> 
>>>>> Forgot to include team.
>>>>> 
>>>>> ---------- Forwarded message ----------
>>>>> From: Ravi Kumar Singh <rk8000 at g.rit.edu <mailto:rk8000 at g.rit.edu>>
>>>>> Date: Wed, Apr 13, 2016 at 2:11 AM
>>>>> Subject: Re: [ndnSIM] Seeking help regarding ndnsim; intreset with different sequence number
>>>>> To: Lixia Zhang <lixia at cs.ucla.edu <mailto:lixia at cs.ucla.edu>>
>>>>> 
>>>>> 
>>>>> Thanks Lixia! I will try setting  up a second consumer to request the same data. 
>>>>> But, in our experiment, we need to see what happens when the same consumer requests the same interest multiple times. I tried writing my own implementation of a consumer, but still only the first interest is working well and rest all interest are not even forwarding from the consumer(same) itself. 
>>>>> 
>>>>> Following is a log of its operation ( with my own implemention of one consumer sending three same interest):
>>>>> 
>>>>> 1: 1s 0 CustomApp:SendInterest(): [DEBUG] Sending Interest packet for /root/%FE%00?ndn.InterestLifetime=1000&ndn.Nonce=4231791817
>>>>> 2: 1s 0 CustomApp:SendInterest(): [DEBUG] Sending Interest packet for /root/%FE%00?ndn.InterestLifetime=1000&ndn.Nonce=2837832477 <tel:2837832477>
>>>>> 3: 1s 0 CustomApp:SendInterest(): [DEBUG] Sending Interest packet for /root/%FE%00?ndn.InterestLifetime=1000&ndn.Nonce=2794407255
>>>>> 4: 1.01002s 1 ndn.cs.Lru:Lookup(0x1493c10, /root/%FE%00)
>>>>> 5: 1.02004s 2 ndn.Producer:OnInterest(0x14fce90, 0x1528ae8)
>>>>> 6: 1.02004s 2 ndn.Producer:OnInterest(): [INFO ] node(2) responding with Data: /root/%FE%00
>>>>> 7: 1.03089s 1 ndn.cs.Lru:Add(0x1493c10, /root/%FE%00)
>>>>> 8: 1.04174s 0 CustomApp:OnData(): [DEBUG] Receiving Data packet for /root/%FE%00
>>>>> 
>>>>> As, you can see from the log, only first interest is working file and rest 2 interest are not even forwarding from the consumer itself.
>>>> 
>>>> Hi Ravi,
>>>> 
>>>> What you see is correct and intentional behavior of an NDN router.  You're trying to send 3 interests for the same data at the exactly the same time.  As a result, these 3 interests will be aggregated into one and only one will be sent further. Also, given that the same app (though the same face) is sending those interests, only one data is returned back.
>>>> 
>>>> Depending on what specific effect you want to observe, you may want to spread out sending interests in time and/or request from different simulated nodes (different app instances).
>>>> 
>>>> ---
>>>> Alex
>>>> 
>>>>> Please suggest!
>>>>> 
>>>>> Regards,
>>>>> Ravi
>>>>> 
>>>>> On Wed, Apr 13, 2016 at 12:31 AM, Lixia Zhang <lixia at cs.ucla.edu <mailto:lixia at cs.ucla.edu>> wrote:
>>>>> 
>>>>>> On Apr 12, 2016, at 9:20 PM, Ravi Kumar Singh <rk8000 at g.rit.edu <mailto:rk8000 at g.rit.edu>> wrote:
>>>>>> 
>>>>>> Hi Team,
>>>>>> 
>>>>>> I am a Graduate student of Rochester Institue of Technology, New York. Currently, I am working on a research paper based on NDN along with my Prof.
>>>>>> Currently, we want to simulate the NDN with your software NDNSIM. We want to simulate the scenario where the consumer is sending the same interest (same sequence number) multiple times to the producer so that the cache Server would serve the interest as it would have the content in its cache instead of serving the request from the producer.
>>>>>> We tried googling but couldn't found any relevant link for this scenario.
>>>>>> 
>>>>>> Requesting you to please help us on how we can simulate the scenario where the consumer is sending the same interest(same sequence number) instead of the incremental sequence number (default behaviour).
>>>>> 
>>>>> it seems to me easier to set up a second consumer to request the same data, rather than forcing the same consumer to send repeated interest.
>>>> 
>>>> 
>>>> _______________________________________________
>>>> ndnSIM mailing list
>>>> ndnSIM at lists.cs.ucla.edu <mailto:ndnSIM at lists.cs.ucla.edu>
>>>> http://www.lists.cs.ucla.edu/mailman/listinfo/ndnsim <http://www.lists.cs.ucla.edu/mailman/listinfo/ndnsim>
> 
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndnsim/attachments/20160424/a3d15553/attachment.html>


More information about the ndnSIM mailing list