[Ndn-interest] NDNVideo: Available over NFD?

Gene Tsudik gts at ics.uci.EDU
Sun Mar 29 17:35:50 PDT 2015


Alex,

If you look closely at the referenced paper, you will see that it *indeed*
shows how to do secure cut-through fragmentation and re fragmentation.
"Secure" means that both hash and signature verification in routers is
supported.
Gene


On Mar 29, 2015 6:24 PM, "Alex Afanasyev" <alexander.afanasyev at ucla.edu>
wrote:

> In my opinion, cut-through forwarding of segments is a possible
> optimization of hop-by-hop fragmentation which can be applied only for some
> of the data packets: only those that are not requested with security
> constraints (e.g., when neither interest name doesn’t contain implicit
> digest nor interest doesn’t include data packet signature constraint).
> Also, cut-through optimization is meaningful when router connect networks
> with similar (same) MTU characteristics.
>
> ---
> Alex
>
> On Mar 29, 2015, at 4:28 PM, GTS <gts at ics.uci.edu> wrote:
>
>  I disagree with the categorical statement below. Hop-by-hop
> fragmentation is indeed an option for NDN, but it's *not* the only one.
> One way to do secure cut-through (end-based and intermediate)
> (re-)fragmentation is described here:
> http://arxiv.org/abs/1405.2861
>
> Gene
>
>
>
> On 3/29/15 5:23 PM, Alex Afanasyev wrote:
>
>  ....
>  maybe a typo?
> that ought to be/really means “hop-by-hop fragmentation and reassembly”
>
>
>  yes. this is a typo that we didn’t catch before.  The conclusion meant
> to highlight that *hop-by-hop* fragmentation is the only viable option for
> NDN.
>
>> A
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndn-interest/attachments/20150329/2214cc3b/attachment.html>


More information about the Ndn-interest mailing list