Hi Kwang-koo, Thanks very much for your comments. My responses are in-line.
On Mon, Jul 29, 2013 at 7:54 AM, KwangKoog Lee <[email protected]>wrote: > Hi. This is Kwang-koog from kt which is one of major network providers in > South Korea. > First, I appreciate a vast of contributions by i2rs members for enabling > intelligent and simple routing operations. > > As Alia mentioned in her draft, many network providers including kt > today have very huge and complex routing networks > according to the growth of the IP-based services but they really need > simple operation and new business model for covering a number of use cases > on that. > With the benefit of the distributed routing manner, modern routing systems > easily plugged into a designated routing network and > forward user packets to desired destinations. But, in the operator > perspective, policy-based routing control and necessary optimizations for > routing are always major issues for them because it is very hard to control > multi-vendor ruters with their own proprietary functions. > > Our company has operated huge routing systems with the size of thousands > of routing systems during tens of years > and made a significant effort to optimize routing systems. > So, current routing systems support preferable service level aggrement and > traffic utilization is optimized by the our own planning rules. > However, to this end, there have been even much trial-and-error and many > times. > So, I expect that the works of i2rs help network providers to easily > control their network and also support many usecases from our customers. > [Alia] I would welcome your thoughts and feedback on the use-cases that I2RS is currently discussing. Additional use-cases would also be quite welcome (given that they fit in the charter) to help scope what information in the models are needed. > > I also think Alia clearly stated such concerns of modern routing > systems so that I also support the WG adoption. > I2RS members pointed out many comments and feedbacks for enhancing that > document which I also agree with. > > In addition, I give a small comment to her draft for "Sec. 5 desired > aspect of a protocol for i2rs" > That is "*interoperability between protocol versions*". > For example, newly defined software define network protocols such as > openflow rapidly changed and a bunch of functions are added. > But, nobody does consider the interoperability of such protocols. > I think providers are very wonderring that different version protocols can > be inter-working or not. > So, I want to comment that i2rs protocol has to have this aspect. > [Alia] A good point. I've added: "Extensibility and Interoperability: Both the I2RS protocol and models must be extensible and interoperate between different versions of protocols and models." Thanks, Alia > Thanks. > > sincerly, > Kwang-koog Lee (Ph.D) > the advanced institue of technology of kt, South Korea > > > On Mon, Jul 29, 2013 at 7:52 PM, Jon Mitchell <[email protected]>wrote: > >> On 24/07/13 17:53 -0400, Alia Atlas wrote: >> > Please review draft-atlas-i2rs-problem-statement-01 and comment on >> whether >> > it should be adopted by I2RS. Detailed technical conversation is also >> most >> > welcome. >> > >> >> Support adoption.. some small issues to authors direct. >> >> -Jon >> >> _______________________________________________ >> i2rs mailing list >> [email protected] >> https://www.ietf.org/mailman/listinfo/i2rs >> > >
_______________________________________________ i2rs mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2rs
