Victor, >-----Original Message----- >From: Victor Pascual ávila Sent: Sunday, March 30, 2008 2:03 AM >To: Henry Sinnreich >Cc: Song Yongchao; P2PSIP Mailing List >Subject: Re: [P2PSIP] P2PSIP diagnosis > >Song, > >On Sat, Mar 29, 2008 at 6:26 PM, Henry Sinnreich <[EMAIL PROTECTED]> wrote: >> This is an interesting and useful method, though I believe the I-D >> should have a section "Usage Scenarios" to make it clear why and when >> p2p network diagnostics are required. > ><keeping p2psip clients in mind> >IMO, different application scenarios (degree of heterogeneity, node >mobility and so on) will require different numbers of peers (few, >many). The point is to design a peer/client turning criteria >(performance, operator policy, etc) which optimizes the performance >while it prevents free-riding
I think this turning criteria is needed. IMO, there will not be one general criteria for it. It will be different for various application scenarios. >Could this diagnosis be useful to evaluate the performance and serve >as an input for the turning criteria? Yes, with the diagnostics, you can evaluate the average performance of the overlay through the diagnostic information gathered from different peers at different time (Traceroute mode of Echo request can get all the diagnostic information of the intermediate peers). By comparing with the diagnostic information from the specific peer/client (Ping mode of the Echo request can be used to get only the diagnostic information of the destination peer), and then you can use the turning criteria to make the decision. -BR -Song >Cheers, >-- >Victor Pascual ávila _______________________________________________ P2PSIP mailing list [email protected] https://www.ietf.org/mailman/listinfo/p2psip
