Re: [p2prg] RGLC review for draft-kamei-p2p-experiments-japan-06

["Tsuyoshi Momose (tmomose)" <[e-mail]>]

Vijay, 

Thanks again.


On 2012/07/12, at 1:19, Vijay K. Gurbani wrote:

> I only have one further comment in -07, which is detailed below.
> 
>>> - S4, second paragraph: You write --- "When a peer joins the
>>> network, it registers its location information (IP address) and
>>> supplementary information (line speed, etc.) with the hint server."
>>> Here, I think you imply that a "peer" is one of your dummy servers,
>>> hence it is happy to provide supplementary information like line
>>> speed, etc. Since you did not modify the real peers, they will not
>>> provide the same supplementary information to the hint server that
>>> your dummy peer will.  Please clarify.
>> 
>> Dummy nodes are used only for measurement. They do not control the
>> p2p network at all. But to make it clear, we updated the document.
> 
> I remain unsure on how a non-dummy peer sends information to a "hint
> server".  My assumption is that a non-dummy peer was not modified, so
> why would it contact the hint server. which appears to be a piece of
> your data-gathering machinery and would not normally appear in the
> swarm?

I may not understand your question correctly, but let me summarize again the purpose and behavior of dummy nodes.
The experiment coordinator put a hint server which behaves like an ALTO server.
Every P2P nodes provided by vendors who join this experiment always use the hint server. The protocol for the hint server is open to the experiment attendee, thus, the vendors can implement the code for the hint server by themselves and they can choose best peers to reduce the traffic. But there are neither methods nor protocols to measure the effect of reduced traffic because it's hard to make standards for quite different p2p applications.
So, to measure the effect, we introduced nodes which dedicate p2p traffic measurement. The nodes build another p2p network system but do nothing except measuring traffic. So, they are called 'dummy nodes'. Dummy nodes just monitors the traffic and analyze their peers at their network interface. 

I hope this summary helps you to understand the situation.



> It may be that I am missing something minor here.
> 
> Apart from the above, I have no further comments on -07.
> 
> I suspect that edits for language etc. will help the draft, but maybe
> these could be done by the RFC editor?

we hope so, but I don't know the further plans.


> Thanks for attending to my review.


We appreciate your sincere review.

Regards,

--
Tsuyoshi Momose