There is a display problem in the last email, which has been corrected as follows:
> Hi, > > As about how to implement inter-domain SAVNET mechanism, it seems that we have several choices, namely, adding a new SAFI, adding a new message type, or implementing a new protocol other than BGP. > > We think the final choice depends on how much overlap there is between BGP protocol messages and the SAV messages. To enable this kind of inter-domain SAV functionality, It is inevitable to introduce additional overhead. But generally we think the cost of adding a new SAFI is less than the cost of adding a new message type and even less than the cost of introducing a new protocol. > > Best, > Lancheng > > -----原始邮件----- > 发件人: "Lancheng Qin" <[email protected]> > 发送时间: 2022-05-09 14:33:16 (星期一) > 收件人: [email protected] > 抄送: "'Robert Raszuk'" <[email protected]>, [email protected], 'RTGWG' <[email protected]> > 主题: Re: RE: [savnet] Regarding reusing existing routing protocols for SAV//RE: SAVNET WG charter > > Hi, > > As about how to implement inter-domain SAVNET mechanism, it seems that we have several choices, namely, adding a new SAFI, adding a new message type, or implementing a new protocol other than BGP. > > We think the final choice depends on how much overlap there is between BGP protocol messages and the SAV messages. To enable this kind of inter-domain SAV functionality, It is inevitable to introduce additional overhead. But generally we think the cost of adding a new SAFI < the cost of adding a new message type < the cost of introducing a new protocol. > > Best, > Lancheng > > > > -----原始邮件----- > > 发件人: [email protected] > > 发送时间: 2022-05-06 21:24:02 (星期五) > > 收件人: "'Robert Raszuk'" <[email protected]> > > 抄送: 'Xingang' <[email protected]>, "'Lancheng Qin'" <[email protected]>, [email protected], 'RTGWG' <[email protected]> > > 主题: RE: [savnet] Regarding reusing existing routing protocols for SAV//RE: SAVNET WG charter > > > > > > > My take is that if existing RFC8955 does not provide a sufficient > > > placeholder to propagate SAV data it should be transported outside of > > > BGP entirely. > > > > I would prefer this information be transported outside BGP entirely, but _if_ it's going to be transported in BGP, please let's do it in a way that reduces the load on existing BGP implementations to a minimum. > > > > The amount of churn represented by this kind of work is probably going to warrant some new way of managing this information anyway. > > > > 😊 /r > </[email protected]></[email protected]></[email protected]></[email protected]> </[email protected]></[email protected]></[email protected]> _______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
