Re: [DNA] Considerations for DNA Schemes with multiple Interface andLayer 2 Technologies

[Greg Daley <greg.daley@eng.monash.edu.au>]

Hi Youn-Geun,

YGHONG wrote:
> Hi..Greg..
> 
> My answers are in lines..
> 
> 
>>-----Original Message-----
>>From: Greg Daley [mailto:greg.daley@eng.monash.edu.au] 
>>Sent: Tuesday, October 19, 2004 11:34 AM
>>To: yghong@etri.re.kr
>>Cc: dna@eng.monash.edu.au; ???; ???
>>Subject: Re: [DNA] Considerations for DNA Schemes with 
>>multiple Interface and Layer 2 Technologies
>>
>>
>>Hi Youn-Geun,
>>
>>I have some questions.
>>
>>Yong-Geun Hong wrote:
>>
>>>Dear DNA Group..
>>>
>>>We submitted the following draft to the IETF Secretariat.
>>>It should appear in the internet-drafts in the internet-drafts 
>>>directory soon.
>>>
>>>"Considerations for DNA Schemes with Multiple Interfaces  
>>
>>and Layer 2 
>>
>>>Technologies"
>>>http://pec.etri.re.kr/~yghong/draft-hong-dna-if-l2-00.txt
>>>
>>>Currently, we are developing multiple interfaces node. 
>>>(one is 802.11g, the other is CDMA2000 1x) While we are developing, 
>>>current DNA schemes are enough for a single interface and  WLAN.
>>>When we consider multiple interfaces and other link-layer 
>>
>>technology, 
>>
>>>we think that current DNA schemes may be insufficient. In this 
>>>document, we consider and analyze various environments for applying 
>>>DNA schemes.
>>>
>>>Please let us know your comments/thoughts.
>>
>>in section 3.3 (page 4) there's a statement:
>>
>>
>>      If each link-layer technology of each interface is the same and
>>      each interface is in the same link, this scenario may 
>>not be very
>>      much different with case 3.1. Because each interface is 
>>in the same
>>      link, link information on each interface may be the 
>>same. In this
>>      scenario, multiple interfaces do not have any effect on 
>>detecting
>>      any link changes.
>>
>>What do you mean by technology here? MAC/PHY combination or 
>>something else?
>>
> 
>   => MAC/PHY combination
> 
> 
>>It may be that two wireless devices have the same MAC 
>>capabilities and same PHY encoding, but have different 
>>sensitivities, policy, antennae (antennas), and physical locations.
>>
>>In this case, a condition affecting the IP layer connectivity 
>>of one may not affect another.  It may actually be why you'd 
>>want to have both interfaces.
>>
> 
>   => You are right. In this case, a condition affecting the IP layer
> connectivity
>     of one may not affect another. But the important thing is that each
> interface
>     is in the same link. Although, two wireless devices have different
> sensitivities,
>     policy, et al, access point(s) which two wireless devices are connected
> to can 
>     be in the same link.
>   => I think that this case may not be common but we can guess this
> situations. 
> 
> 
>>Do you have any papers or tech-reports which illustrate the 
>>situation as you described?
>>
> 
>  => No..As we are designing multiple interface node, we have consider this
> case.
> 
>>There are a lot of statements about connection of another L2 
>>medium, but limited correlation of these processes to DNA.
>>I think that if there's work needed for CDMA systems, it may 
>>be necessary to be specific as to the actual issues.
>>
>>Is it possible to work out a succinct problem statement for 
>>these issues?
>>
> 
>   => What I want to say in this section is that in this case if each
> interface is in the same link,
>      the effect of multiple interfaces for DNA schemes may be none.

It would be good to have some corroboration of this.
Anecdotally things may seem right, but perhaps there is an
underlying issue here which we need to expose.

Greg