Re: [P2PSIP] Re: HIP pros and cons

[Philip Matthews <[e-mail]>]

It is true that the definition of "p2p layer" has been vague.
This reflects the uncertainty in the WG about how we want to  
structure our work.

When draft-matthews-p2psip-hip-hop-00 was written, it was not yet  
clear that the WG was going to split SIP from the p2p layer. Hence  
the use of that term in the draft.

However, I think many people these days use "p2p layer" as roughly a  
synonym for "Peer Protocol".

- Philip

On 16-Dec-07, at 21:58 , Wei Gengyu wrote:

> Henry,
>
> I am very appreciated your comments. And no new questions about  
> this thread.
>
> But I hope to get hints on some already raised questions.
> Would you please help me clarify that?
>
> What is p2p layer? What do you mean "p2p layer"?
>
> In "draft-matthews-p2psip-hip-hop-00.",we can see 'p2p-layers'
> that including IPv4 or IPv6, UDPv4 or UDPv6?, HIP or ESP, TCP or  
> UDP, and distributed database.
> The 'p2p layers'contain five layers.
>
> I need to know the difference between "p2p layer" and 'p2p layers'.
> So, I shall see where you put HIP under p2p layer.
>
> In "draft-hautakorpi-p2psip-with-hip-01.txt", there are four  
> suggestions in protocol layer scheme.
> Only (a) of Figure 3 contains HIP, but HIP is set on top of Peer  
> protocol.
> whist there is no words of "p2p layer", it seems that Peer protocol  
> should at that layer.
>
> Refer to  "draft-willis-p2psip-concepts-04 - Concepts and  
> Terminology for Peer to Peer SIP",
> "2.  High Level Description
>
>    A P2PSIP Overlay is a collection of nodes organized in a peer-to- 
> peer
>    fashion for the purpose of enabling real-time communication  
> using the
>    Session Initiation Protocol (SIP).  Collectively, the nodes in the
>    overlay provide a distributed implementation of the location  
> service
>    [RFC3261] for mapping Addresses of Record (AoRs) to Contact URIs.
>    They also provide a transport service by which SIP messages can be
>    transported between any two nodes in the overlay.
>
>    A P2PSIP Overlay consists of one or more nodes called P2PSIP Peers.
>    The peers in the overlay collectively run a distributed database
>    algorithm.  This distributed database algorithm allows data to be
>    stored on peers and retrieved in an efficient manner.  It may also
>    ensure that a copy of a data item is stored on more than one  
> peer, so
>    that the loss of a peer does not result in the loss of the data  
> item
>    to the overlay.   "
>
> Unforunately, there is no explicit definition of "p2p layer" in the  
> I-D
> although so many people say "p2p layer" in this mailing list.
>
> Even it seems to be a silly quetion, "p2p layer" is still a vague  
> concept when people say it.
> So, I think that WG needs to make this basic definition clear.
>
> Best regards,
>
> Gengyu
> ----- Original Message -----
> From: Henry Sinnreich
> To: Wei Gengyu ; P2PSIP Mailing List
> Sent: Monday, December 17, 2007 7:21 AM
> Subject: RE: [P2PSIP] Re: HIP pros and cons
>
> > My problem is when HIP is used at the application layer,
>
> > or using the same algorithm to generate Peer ID.
>
>
>
> HIP runs below the application layer and also below the p2p layer.
>
> HI is different from the p2p nodeID or application layer (such as  
> SIP) identifiers, such as AoR.
>
>
>
> Henry
>
> From: Wei Gengyu [mailto:[e-mail]]
> Sent: Sunday, December 16, 2007 5:25 AM
> To: P2PSIP Mailing List
> Subject: Fw: [P2PSIP] Re: HIP pros and cons
>
>
>
>
>
> ----- Original Message -----
>
> From: Wei Gengyu
>
> To: [e-mail] ; [e-mail] ; Philip  
> Matthews
>
> Cc: P2PSIP Mailing List
>
> Sent: Saturday, December 15, 2007 10:25 AM
>
> Subject: Re: [P2PSIP] Re: HIP pros and cons
>
>
>
> Jeff,Spencer, and Philip,
>
>
>
> First, thank you all for your correction.
>
>
>
> HIP might work well as RFC4423 defined.
>
>
>
> My problem is when HIP is used at the application layer,
>
> or using the same algorithm to generate Peer ID.
>
>
>
> If HIP-like algorithm is used in the overlay while HIP is used  
> between network layer and transport layer,
>
> the Peer ID will share the same name space with Host ID.
>
> For rfc4423, when a node have multiple Host IDs, they only cost  
> memory spaces a little.
>
> If one host are permited to have multiple Peer IDs that happen to  
> belong to one overlay,
>
> it would incur potential risks to the P2PSIP overlay.
>
>
>
> And it seems not be capable to tackle this case in RVS of HIP.
>
> Is there something wrong?
>
>
>
> Regars,
>
>
>
> Gengyu
>
> _______________________________________________
> P2PSIP mailing list
> [e-mail]
> https://www1.ietf.org/mailman/listinfo/p2psip

_______________________________________________
P2PSIP mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/p2psip