Re: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"

["Ji-In Kim" <[e-mail]>]

Dear Jong-Hyouk Lee,

Thank you for the review.

There is an issue on how to assign the HNP in DMM solution. It needs more discussion.
However, if the HNP is not assigned to the PMIP domain that consist of several MAGs, all of MAGs in the Internet shall join the multicast group for control operations.
In this case, there may be serious multicast overhead.
(In SD-PMIP, the multicast query operations will be performed to all of MAGs.)



 

Best Regards, Ji-In Kim

----------------------------------------------------------------------

Ph.D student, Communications Protocols Lab, Kyungpook National University.

1370 Sankyuk-dong, Buk-gu, Daegu, Korea.

E-mail: [e-mail] 
Tel: [phone] C.P: 82-10-9371-1324

 

From: Seok-Joo Koh [mailto:[e-mail]] 
Sent: Friday, December 09, 2011 7:47 PM
To: [e-mail]
Subject: Fwd: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"

 

 

 

**************************

Seok-Joo Koh

[e-mail]

**************************


전달된 메시지 시작:

보낸 사람: Seok-Joo Koh <[e-mail]>
날짜: 2011년 12월 9일 오후 7시 5분 58초 KST
받는 사람: Ji-In Kim <[e-mail]>
참조: [e-mail]
제목: Re: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"
답장받는 사람: [e-mail]

수정사항..

 

1. Thank you for the review. Please see inline.

--> See inline 생략 (이건, 문장 중간에 답변을 달때 사용하는 말임)

 

2. There is an issue how to assign the HNP in DMM.

--> There is an issue on how ~

 

3. It needs to discuss.

--> It needs more discussion


4. However, if the HNP is not assigned to the PMIP domain to consist of several MAGs, all of MAGs in the Internet join the multicast group for control signals.

--> However, if the HNP is not assigned to the PMIP domain that consists of several MAGs, all of MAGs in the Internet shall join the multicast group for control operations.

 

5. In this case, there is serious multicast overhead.

--> In this case, there may be serious multicast overhead.

 

6. (As connection is established, multicast query operations perform to all of MAGs.)

--> (In SD-PMIP, the multicast query operations will be performed to all of MAGs.)

 

 

2011년 12월 8일 오후 8:59, Ji-In Kim <[e-mail]>님의 말:

안녕하세요. 교수님 지인입니다.

아래와 같이 답변을 할 생각입니다.

Dear Jong-Hyouk Lee,

Thank you for the review. Please see inline.

There is an issue how to assign the HNP in DMM solution. It needs to discuss.
However, if the HNP is not assigned to the PMIP domain to consist of several MAGs, all of MAGs in the Internet join the multicast group for control signals.
In this case, there is serious multicast overhead.
(As connection is established, multicast query operations perform to all of MAGs.)

Multicast에 대한 overhead가 HNP 할당으로 인해 줄어들 경우 Multicast 방식이 좀 더 DMM가 가까운 어프로치가 될 것 같아 제외할 계획이라는 말은 일단 제외하였습니다.
문서 업데이트 시 SD-PMIP에 대한 핸드오버 시나리오도 고려를 해서 포함해야 할 것 같습니다.


김지인 드림
----------------------------------------------------------------------
경북대학교 IT대학 컴퓨터학부 통신프로토콜 연구실 박사과정
경북대학교 IT대학 3호관 309A호
E-mail: [e-mail]
Tel: [phone] C.P: 010-9371-1324


-----Original Message-----
From: Seok-Joo Koh [mailto:[e-mail]]
Sent: Wednesday, December 07, 2011 9:21 PM
To: Ji-In Kim
Subject: Re: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"

오케이,
텍스트 준비해서 보내주삼..


**************************
Seok-Joo Koh
[e-mail]
**************************

2011. 12. 7. 오후 8:05 "Ji-In Kim" <[e-mail]> 작성:

> 안녕하세요. 교수님 지인입니다.
>
> 여러모로 생각을 해 본 결과 해당 질문에 대한 대답을 아래와 같은 방향으로 작성해야 할 것 같습니다.
>
> HNP를 어떻게 할당할지에 대해서는 현재 DMM에서 충분히 이슈가 되고 있으며, 아직까지 논의를 해야 할 부분이 많다.
> 하지만 PMIP 도메인 별로 HNP로 묶어서 통신을 하지 않을 경우, Internet 상에 있는 모든 MAG들이 멀티캐스트 그룹에 참여를 해야 하고, 이로 인한 Multicast overhead가 너무나 심각하다.
> (세션이 하나 열릴 때 마다 모든 MAG에 Multicast Query를 수행해야 한다.) 하지만 이 경우 본 문서에 있는
> S-PMIP과의 차이점이 모호해지는 문제가 존재를 한다.
> 그래서 현재 SD-PMIP 기법에 대해 제안 여부를 고민하고 있다.
>
> 이렇게 답변을 한다면 다음 문서에서 SD-PMIP을 제외하는 것에 대한 점도 어느 정도 MEXT WG에 어필할 수 있고 답변도 될 것 같습니다.
>
> 교수님 생각은 어떠신지요?
>
>
> 김지인 드림
> ----------------------------------------------------------------------
> 경북대학교 IT대학 컴퓨터학부 통신프로토콜 연구실 박사과정
> 경북대학교 IT대학 3호관 309A호
> E-mail: [e-mail]
> Tel: [phone] C.P: 010-9371-1324
>
>
> -----Original Message-----
> From: Seok-Joo Koh [mailto:[e-mail]]
> Sent: Wednesday, December 07, 2011 8:39 AM
> To: 김지인
> Subject: Fw: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"
>
> PMIP의 HoA (단말)와 PCoA(MAG)를 혼동하고 있는듯 한데..
>
> 어떤 주소에 대한 prefix ??
>
> 응답시에 관련 용어를 분명히 하는게 좋을듯..
>
>
> *************************
> Seok-Joo Koh
> [e-mail]
> *************************
>
> ----- Original Message -----
> From: "Jong-Hyouk Lee" <[e-mail]>
> To: "Ji-In Kim" <[e-mail]>
> Cc: "고석주교수님" <[e-mail]>; "'mext'" <[e-mail]>
> Sent: Wednesday, December 07, 2011 5:08 AM
> Subject: Re: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"
>
>
>> Dear Ji-In Kim,
>>
>> I do not think that all mobile nodes have the same prefix in a SD-PMIP domain consisting of MAGs.
>> I thought that a network prefix assigned to a mobile node is managed
>> by a serving MAG. Are you really considering that all of the nodes in the SD-PMIP domain have the same prefix?
>>
>> Cheers.
>>
>> On Tue 06 Dec 2011 04:01:33 AM CET, Ji-In Kim wrote:
>>>
>>> Dear Jong-Hyuok Lee,
>>>
>>> Thank you for the valuable comments. Please see my inline responses.
>>>
>>> 1) We suppose that both MN and CN are in the same domain in the
>>> S-PMIP and SD-PMIP. So, all of the nodes in SD-PMIP have the same prefix.
>>> 2) In the draft, we didn't consider the handover scenario yet.
>>> However, we will revise the draft to include the handover scenarios within this month.
>>>
>>>
>>> Best Regards, Ji-In Kim
>>> --------------------------------------------------------------------
>>> -- Ph.D student, Communications Protocols Lab, Kyungpook National
>>> University.
>>> 1370 Sankyuk-dong, Buk-gu, Daegu, Korea.
>>> E-mail: [e-mail]
>>> Tel: [phone] C.P: 82-10-9371-1324
>>>
>>>>
>>>> -----Original Message-----
>>>> From: [e-mail] [mailto:[e-mail]] On
>>>> Behalf Of
>>>
>>> Jong-Hyouk Lee
>>>>
>>>> Sent: Monday, December 05, 2011 1:36 AM
>>>> To: [e-mail]; mext
>>>> Subject: [MEXT] Comments on "draft-sjkoh-mext-pmip-dmc-03"
>>>>
>>>> Dear authors,
>>>>
>>>> I read the draft "draft-sjkoh-mext-pmip-dmc-03" and have
>>>
>>> comments/questions:
>>>>
>>>>
>>>> How is an MN's handover handled in SD-PMIP? For instance: 1) It's
>>>> not
>>>
>>> clear how a prefix of the MN is provided; and 2) it is also not
>>> clear from the draft how ongoing packets (destined to the MN or sent
>>> from the
>>>>
>>>> MN) are handled during handovers.
>>>>
>>>> Thanks.
>>>>
>>>> --
>>>> IMARA Team, INRIA, France
>>>> Jong-Hyouk Lee, living somewhere between /dev/null and /dev/random
>>>>
>>>> #email: jong-hyouk.lee (at) inria.fr || hurryon (at) gmail.com
>>>> #webpage: http://sites.google.com/site/hurryon/
>>>>
>>>> _______________________________________________
>>>> MEXT mailing list
>>>> [e-mail]
>>>> https://www.ietf.org/mailman/listinfo/mext
>>>
>>>
>>> --
>>> IMARA Team, INRIA, France
>>> Jong-Hyouk Lee, living somewhere between /dev/null and /dev/random
>>>
>>> #email: hurryon (at) gmail.com || jong-hyouk.lee (at) inria.fr
>>> #webpage: http://sites.google.com/site/hurryon/
>





 

-- 
 
***********************
Seok-Joo Koh
[e-mail]
***********************