Re: Requirements on cmm
jouni korhonen <[email protected]>
| Newsgroups | gmane.ietf.nemo |
|---|---|
| Message-ID | <[email protected]> |
Behcet, On Nov 19, 2012, at 10:41 PM, Behcet Sarikaya wrote: > Hi Brian, > > On Mon, Nov 19, 2012 at 1:53 PM, Brian Haberman > <[email protected]> wrote: >> Behcet, >> >> >> On 11/19/12 12:25 PM, Behcet Sarikaya wrote: >>> >>> Hi all, >>> >>> Here is the requirement I have in mind: >>> >>> The DMM solution SHOULD support more than one cloud network belonging >>> to the same DMM domain. >>> >>> Two points: >>> >>> The above requirement does have protocol impact. >> >> >> Could you elaborate on what that impact would be? >> >> It is unclear to me what the impact of using a cloud service would be. The >> IETF has not had to change other protocols in order to have them work within >> "the cloud". > > I think it is early to say. I am confused. If you think it is too early to say, then how can you claim cloud has an impact? If there is something concrete you have in mind, just say it. - Jouni > > Cloud networks have been all Layer 2 based and because of that many > mobility issues do not surface. For example virtual machine mobility > is just change of Ethernet address and no IP layer changes are needed. > > However, I think that the trend is towards Layer 3 based clouds and > also inter-cloud communication needs are increasing. For example there > are many people advocating the development of an IETF virtual machine > mobility protocol. > > I think we need to look at DMM from this perspective,and see if we can > make our mobility protocols more suitable to the cloud. > > That is the reason why I wrote > draft-sarikaya-dmm-cloud-mm. > With even the simplistic assumptions, my drafts shows that the cloud > does have some protocol implications on our protocols. > > Regards, > > Behcet > _______________________________________________ > dmm mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/dmm