[Ndn-interest] Proposed deprecation: byte offset segmenting

[Mosko, Marc <mmosko@parc.com>]

I think the byte-offeset name indexes were much easier to use with exclusions, as you could pipeline those (i.e. do every 1500 or4000 or 8000 bytes and you'll get the nearest thing).  Without exclusions, I think you are correct that you then need a manifest or some other indication of naming convention, in which case I don't see the value of byte-offset.  At least for off-line publisher.

For on-line publishers, it could be used like the HTTP byte offset to fetch whatever byte range one is interest in.  Though I think that is mostly used to stripe request between different replicas.  Maybe combining that with routing hints would be interesting?  Though one could do pretty much the same think with segment numbers in that case.

Marc

________________________________________
From: Ndn-interest <ndn-interest-bounces at lists.cs.ucla.edu> on behalf of Junxiao Shi <shijunxiao at email.arizona.edu>
Sent: Wednesday, January 30, 2019 5:56:36 PM
To: Nick Briggs
Cc: ndn-interest
Subject: Re: [Ndn-interest] Proposed deprecation: byte offset segmenting

Dear folks

20190130 NFD call discussed this issue.
While there aren’t many use cases, the cost of preserving byte offset segmenting is low.
Therefore, Alex abandoned his plan of deprecating byte offset segmenting.

A use case I can think of is segmenting content at non uniform chunk sizes, e.g. boundary determined by Rabin fingerprints. I had a related project although it does not directly use byte offset segmenting: Content-Addressable NDN Repository https://github.com/yoursunny/carepo
Retrieval of byte offset segmented content would require either stop-and-wait, or a manifest listing all segment boundaries to enable pipelining.

Yours, Junxiao

On Tue, Jan 29, 2019 at 15:28 Nick Briggs <nicholas.h.briggs at gmail.com<mailto:nicholas.h.briggs at gmail.com>> wrote:
What is the benefit of removing it from the specification (or the cost of leaving it) ?

-- Nick


> On Jan 29, 2019, at 12:22 PM, Junxiao Shi <shijunxiao at email.arizona.edu<mailto:shijunxiao at email.arizona.edu>> wrote:
>
> Dear folks
>
> NDN Naming Conventions revision 1 defines two segmenting markers: 0x00 and 0xFB.
> 0x00 is the segment number marker. Segments of a content are assigned consecutive segment numbers.
> It is commonly used in many applications, including ndnputchunks, NDNFS, and PSync.
>
> 0xFB is the byte offset marker. Segments of a content are given this marker followed by payload's byte offset within the content.
> We are unaware of any application that adopts the 0xFB byte offset marker.
> Moreover, unless using a manifest, the consumer cannot predict the byte offset of every segment, and thus hinders the efficiency of content retrieval because pipelining cannot be used.
>
> Alex has promised to discontinue byte offset segmenting in NDN Naming Conventions revision 2.
> If anyone on this mailing list has an application that currently uses 0xFB marker, please speak up within 5 days.
>
> Yours, Junxiao
> _______________________________________________
> 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