Hi Lee, Thanks for discussing technical details. > KRP doesn’t just assume routing follows political geography, it requires it. That’s just terrible network design. Political geography can be added to the IP address assignment, KRP uses the stored information of the KRP RN and KRP ASN only. > NEP has just three metrics: bandwidth (but not capacity), hop count, and delay (RTT?) Can you define capacity? > How does a router know these values for every possible destination address? Every router has a table for every connected interface's bandwidth and delay, and with advertisements, it adds these values in addition to increasing the number of hops to the neighboring router(s) before advertising to have the link total BW, total delay and number of hops. > As a simple exercise, consider what happens when somebody adds or changes a router or link six hops away: how long does it take for all routers to advertise the update and respond to advertisements and advertise updates from that? You mean convergence, if a change to a link or router status occurs, NEP routers recalculates routes and eventually updates the existing routing table, it is like how GPS works, if a route for example has a high traffic due to an accident, it suggests another best route, but with large networks I think NEP needs a faster approach and this is what I'm working on now. Khaled -----Original Message----- From: Lee Howard [mailto:lee@xxxxxxxxxx] Sent: Thursday, December 21, 2017 6:03 PM To: Khaled Omar; ietf@xxxxxxxx Subject: Re: When the IETF can discuss drafts seriously? On 12/21/17, 6:04 AM, "ietf on behalf of Khaled Omar" <ietf-bounces@xxxxxxxx on behalf of eng.khaled.omar@xxxxxxxxxxx> wrote: >Hi Moonesamy, > >> I suggest comparing (published) RFCs on the subject you are >>interested in with the drafts you have submitted to assess the >>technical and editorial level which is being sought. > >>From your experience with RFCs on this topic, do you think there are >>other RFCs discussing the same EGP and IGP suggested on the KRP and >>NEP IDs? I doubt it, because these protocols are not good ideas. KRP doesn’t just assume routing follows political geography, it requires it. That’s just terrible network design. NEP has just three metrics: bandwidth (but not capacity), hop count, and delay (RTT?). How does a router know these values for every possible destination address? But it’s not three metrics, it’s one comprising those three, because you know the exact best mix of metrics for every network. As a simple exercise, consider what happens when somebody adds or changes a router or link six hops away: how long does it take for all routers to advertise the update and respond to advertisements and advertise updates from that? There is nothing salvageable from these proposals: they are fundamentally naive about networking. I can’t even think of what to suggest to you. Lee >> > >Khaled > >-----Original Message----- >From: ietf [mailto:ietf-bounces@xxxxxxxx] On Behalf Of S Moonesamy >Sent: Thursday, December 21, 2017 12:47 PM >To: Khaled Omar; ietf@xxxxxxxx >Subject: Re: When the IETF can discuss drafts seriously? > >Hi Khaled, >At 12:46 PM 19-12-2017, Khaled Omar wrote: >>I noticed that the IETF participants gives only negative comments >>regarding the submitted IDs, that is good in some cases if it is true, >>but to ignore the positive side and the added values on every draft is >>something that should be changed, I always aim to find a true >>technical discussion on the mailing list to add something new or to >>correct something wrong with confidence. > >Several years ago, I rewrote a draft and sent it to the authors as I >estimated that it would be less effort compared to the effort I put in >to follow up on the review of the draft. Is it worth reviewing a draft >if the amount of effort it requires is much more than the effort the >author put it to write the draft? > >I suggest comparing (published) RFCs on the subject you are interested >in with the drafts you have submitted to assess the technical and >editorial level which is being sought. > >Regards, >S. Moonesamy > >