Hi Albert, While it is not unique as you mentioned it is unique in comparison to traditional routing techniques. In addition the ability from the owner of the EID having the ability to specify how load balance/distribute inbound traffic is also a benefit we get from the pull model. I do agree we need to hit all 4 as you mentioned, but most end users who evaluate LISP in my experience usually highlight that the pull model and being able to control traffic inbound is a key reason the decide on LISP. With that said I believe it should be considered for addition in your document based on this.
My 2 cents, Chad Hintz Sent from my mobile device, please excuse the spelling mistakes. > On Oct 12, 2014, at 6:05 PM, Albert Cabellos <[email protected]> > wrote: > > Hi Ronald > > Thanks for your comment. > > I agree that none of such design principles are unique to LISP, but -I > think- you are reading them independently and they should be > considered the four of them *at the same time*. With this I expect > that the reader gets -very quickly- LISP's big picture. > > I don´t think that pull is such a unique characteristic, DNS works > based on exactly the same principle: "pull locators". > > Albert > > > >> On Sun, Oct 12, 2014 at 3:32 AM, Dino Farinacci <[email protected]> wrote: >> Well everything tends to look the same but not in this case. This is the >> first mapping database that is really fully specified and tested at the >> network layer. >> >> Dino >> >> >>> On Oct 11, 2014, at 9:20 PM, Ronald Bonica <[email protected]> wrote: >>> >>> Dino, >>> >>> That too! >>> >>> However, the mapping database system is not entirely unique to LISP. Every >>> architecture that maps one address space to another needs a data base to >>> maintain mapping information. The part that is unique to LISP is how the >>> data is distributed >>> >>> Ron >>> >>> >>>> -----Original Message----- >>>> From: Dino Farinacci [mailto:[email protected]] >>>> Sent: Saturday, October 11, 2014 9:02 PM >>>> To: Ronald Bonica >>>> Cc: [email protected] >>>> Subject: Re: [lisp] draft-ietf-lisp-introduction - Design Principles and >>>> Use >>>> Cases >>>> >>>>> On Oct 11, 2014, at 7:51 PM, Ronald Bonica <[email protected]> wrote: >>>>> >>>>> In Section 2.1, we say that LISP is built on top of four basic design >>>>> principles: >>>>> >>>>> - Locator/Identifier split >>>>> - Overlay architecture >>>>> - Decoupled data and control-plane >>>>> - Incremental deployability >>>> >>>> You left out one that is really important: >>>> >>>> - A Mapping Database System >>>> >>>> Dino >> >> _______________________________________________ >> lisp mailing list >> [email protected] >> https://www.ietf.org/mailman/listinfo/lisp > > _______________________________________________ > lisp mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/lisp _______________________________________________ lisp mailing list [email protected] https://www.ietf.org/mailman/listinfo/lisp
