[Ndn-interest] [Mini-NDN] Measurement Table - new discussion link

[Aman Maldar]

Yes Ashlesh, it was redirected to ndn-interest and not answered.  However,
I have very similar (and kind of rookie) question to the one asked in that
mail.

Repeating the question from previous mail and adding some of my question
later.

>* Is there any way (on current NFD implementation) that one can list the Measurements Table entries using something as nfd-status? Or just accessing via code methods get/setEntry ?
*> >* I also have questions about how it is used by the forwarding
strategies. On NFD-dev guide it states that the strategies use the
table to store arbitrary information (can be delay, jitter, RTT,
etc.), but on http://named-data.net/doc/NFD/current/overview.html
<http://named-data.net/doc/NFD/current/overview.html>
<http://named-data.net/doc/NFD/current/overview.html
<http://named-data.net/doc/NFD/current/overview.html>> it says that
the strategies stores past performance results. How "past" is that?
How often does strategies update this table?
*> >* I am guessing that what you would like to store varies from each
forwarding strategy, but is there any common frequency of updates that
the strategies use to input entries on the table (e.g. every
successful interest-data retrieval)?*


I see a name-tree structure here.    http://irl.cs.ucla.edu/~cawka/
nfd-uml/content/ClassDiagram_YTl_NcKD_AAAARxT.html
Measurement table basically allows accessing/storing pit,fib entries using
get(), findexactmatch(), findlongestprefixmatch() functions.

Additionally, is it possible to store Content store hit time/ hit rate
(assuming I calculate it somehow) in the measurement table? All I am trying
to figure out is, what are the parameters which are already stored in
measurement table and how feasible it is to add new parameters to that
table?

My understanding is, the existing tree implementation does not support
adding a new <parameter, value> pair to measurement table. Am I correct?

If above storing <parameter, value> pair concept is not feasible, do I need
to create a new data structure to store measurement parameters like
hit/miss rate, latency etc?

Thank you for long read and your valuable time.

On Mon, Dec 4, 2017 at 12:49 PM, Ashlesh Gawande (agawande) <
agawande at memphis.edu> wrote:

> Probably the message was redirected from ndn-interest (general queries) to
> nfd-dev (development related queries).
>
> And after that nobody answered.
>
>
> Ashlesh
> ------------------------------
> *From:* Mini-NDN <mini-ndn-bounces at lists.cs.ucla.edu> on behalf of Aman
> Maldar <amanmaldar at gmail.com>
> *Sent:* Saturday, December 2, 2017 10:41:12 PM
> *To:* mini-ndn at lists.cs.ucla.edu
> *Subject:* [Mini-NDN] Measurement Table - new discussion link
>
> Hello,
>
> http://www.lists.cs.ucla.edu/pipermail/nfd-dev/2016-April/001680.html
>
>  I am trying to find the new link to the redirected discussion from the
> link above. However, I didn't find anything in archive .
> Can someone please bring up the next thread?
>
> Thanks.
>
>
> --
> *Regards,*
> * Mr. Maldar Aman M.*
> *University of Massachusetts, Lowell*
>



-- 
*Regards,*
*Mr. Maldar Aman M.*
*University of Massachusetts, Lowell*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndn-interest/attachments/20171204/83b6fe28/attachment-0001.html>