[Nfd-dev] Multicast ad-hoc enable and broadcast storm

[Mahyuddin Husairi]

Thanks for the clarification.
Will do some experiment to examine how good current interest suppression
mechanism in nfd to avoid broadcast storm in wireless ad-hoc network. My
initial experiment shows some promising result. The problem is there is
lack of literature for me to cite something related to current interest
suppression mechanism in nfd.

On 15 Aug 2017 21:04, "Lan Wang (lanwang)" <lanwang at memphis.edu> wrote:

I’m not sure exactly what specific broadcast storm scenario you are
referring to, but the exponential suppression implemented with the
multicast strategy will make sure that an interest of the same name
received from the same face won’t be forwarded within an initial window of
10ms.  If more of the same interests are received within the window then
the suppression window will increase exponentially.  This behavior should
prevent the wireless nodes from forwarding this interest repeatedly to each
other (although it is probably not the original intention of this mechanism
as it was designed before the ad hoc mode was added).

Lan

On Aug 14, 2017, at 5:23 PM, Mahyuddin Husairi <mahyuddin at gmail.com> wrote:

Thanks for a reply Prof. Lan.
What i understand from nfd dev guide and nfd source codes, interest
suppression mechanism main function is to protect the network from interest
DDOS attack. Does the interest suppression mechanism is good enough to
prevent wireless network from broadcast storm (with multicast ad-hoc
setup)?
Is there any technical report or literature that discuss about interest
suppression algorithm implemented in nfd?

On Mon, Aug 14, 2017 at 11:43 PM, Lan Wang (lanwang) <lanwang at memphis.edu>
wrote:

> The multicast strategy has an interest suppression mechanism
> https://redmine.named-data.net/issues/2062.
>
> Lan
>
> On Aug 14, 2017, at 4:20 AM, Mahyuddin Husairi <mahyuddin at gmail.com>
> wrote:
>
> AFAIK, current NFD can allow multicast ad hoc through both udp and raw
> ethernet instead suppressed in forwarding pipeline like before. My concern
> is, is there any mechanism, probably in interest loop pipeline to avoid
> broadcast storm happen during multicast in ad-hoc WiFi mode?
>
> --
> Human knowledge Belongs to The World
> _______________________________________________
> Nfd-dev mailing list
> Nfd-dev at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev
>
>
>


-- 
Human knowledge Belongs to The World
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20170815/54db4a33/attachment.html>