[Ndn-interest] Questions about CanBePrefix

[김학서]

Dear Alex,

Thank you very much for you reply.

HS Kim





2019. 5. 2. 오후 9:27, Alex Afanasyev <cawka1 at gmail.com<mailto:cawka1 at gmail.com>> 작성:


On May 2, 2019, at 12:33 AM, 김학서 <tuple at etri.re.kr<mailto:tuple at etri.re.kr>> wrote:

Dear ALL,

I’m reading the NDN Packet Format v0.3 Transition Plan. In the CanBePrefix Switchover section, it is described as follows.

=============
This step discourages the use of CanBePrefix=1 to improve network performance.
=============

My Questions are :

1. What intention do you use the “discourages” in the sentence above? Could you elaborate a bit?

This is not exactly a correct word. What is meant here is that the default setting when the flag is not set is to treat the Interest as a request for the exact name data.  And the whole point is to emphasize that library setting switch to follow the semantics of the packet spec.


2. Can I use the CanBePrefix flag in my new applications?

Yes


3. You mentioned performance degradation in the sentence above, do you have any test results about performance of CanBePrefix? If you have it, can I get it?

It is based on a simple fact that finding data by exact name is simpler (1-step process, if considering hash table implementation) than finding data that matches interest prefix (multiple matches).

—
Alex



Best Regards,
HS Kim
_______________________________________________
Ndn-interest mailing list
Ndn-interest at lists.cs.ucla.edu<mailto:Ndn-interest at lists.cs.ucla.edu>
http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndn-interest/attachments/20190509/2edef9de/attachment.html>