Re: [DNA-BOF] Re: [mobile-ip] Announce: BoF Proposal: DetectingNetwork Attachment

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

G'Day JinHyeock,

Glad you're on board.

JinHyeock Choi wrote:
> Hi
> 
> Thanks for nice work. BoF description seems reasonable to me. 
> 
> Kindly finds my in line comments.
> 
> 
>>Detecting Network Attachment (DNA) Proposed BoF Description:
> 
> 
> The notion of 'Detecting Network Attachment' is not clear to me. 
> What is the difference with 'Movement Detection'? 

I think that Detecting network Attachment is supposed
to have function which may be useful for IP mobility,
but also which is useful in other situations,
such as hosts receiving multicast streams which
may not require mobility signalling.

The idea is to see if there is a common platform
for attachment detection which may be used by other
technologies. I think there has been
discussion indicating that each of IPv4 and IPv6
would have different requirements though.

In this way, only one set of (for example) router
discovery messages will be sent for IPv6.
The results of the detection would then be available
can provide information to MobileIP, ND, Address
Configuration and Upper Layer (TCP/UDP) subsystems.

> 
>>Network Attachment occurs when a host arrives on a new
>>IP subnet.  When attaching to a network, a host either
>>already has a valid configuration for this subnet or
>>must configure addresses.  A host determines whether
>>it requires additional configuration by Detecting Network
>>Attachment.
>>
>>When a host has existing upper layer protocols sessions,
>>it is important to receive a timely indication that
>>attachment has occurred.  This may be the case if a host
>>is connected intermittently, is moving or has urgent data
>>to transmit upon attachment to a link.
>>
>>For these nodes, it is also important detect if an acquired
>>link is new, or has already been visited. This information
>>may be used to distinguish between events where
>>configuration must be initiated, or a host already has
>>valid configuration.
>>
>>This meeting hopes to providing a forum for those
>>interested in developing generic attachment detection
>>technologies for IPv4 and IPv6.
>>
>>The BOF aims to:
>>
>>* Describe existing issues encountered in DHC, ZEROCONF
>>   and Mobileip WGs, which could benefit from work on
>>   detecting network attachment.
>>
>>* Define the problem scope, and environments where network
>>   attachment detection is desirable.
>>
>>* Determine if sufficient interest exists to form a
>>   Working Group on this topic.
>>
>>* Reach consensus on the area of work for a potential WG,
>>   including which problems are outside scope.
> 
> 
> I am not familiar with the issues from DHC, ZEROCONF. But with 
> respect to mobileip, I wish that the work area includes the following. 
> (I use the term MD because the notion DNA is still unclear to me.)
> 
> 1. The issues and problems of MD 
> - The definition (and tasks) of MD 
> - The effect of wireless environment on MD
> - The inconsistency of RA information
> - The delay from (NUD like) reachability test and its effects on performanc

I think that analysis of the existing systems are
important although Internet area ADs pointed out
that there was overlapping work on this from
different groups.

It may be that we can solve the issues for
all of our interesting sub-systems.  This may
mean revisiting analysis which yourself and
others have done on MD, to see if it is applicable
to indicating address changes for example to the
host's DHCPv6 subsystem.

Certainly we need to discuss wireless link issues
for DNA, although we should be careful not to duplicate
work undertaken by PILC WG.

For the issue of delay's effects on upper layer protocols,
I think that this should be handled either
a) as a driver/motivation for using DNA's techniques
b) if we have a base (slow) detection mode, then it may be
    worth describing the requirements for faster detection.

> 2. Basic MD scheme
> - a default MD scheme on which vendors can safely fall back.
> - Eager Cell Switching with L2 trigger 
> - NS/ NA (RS/ RA) exchange 

I think that the work which has been done on
describing current issues in MD may be useful
for a more generic DNA system.
For one thing, movement detection is now
beginning to be understood well.

> 3. Optimized MD schemes
> - Optimized MD scheme which can support seamless handoff
> - Quick Router Discovery: FastRA, FRD
> - LinkID
> - RA optimized for MD

I'm not sure how to proceed on these optimizations.
They are sure to have different applicability in
certain conditions (so I wouldn't favour one or another).

Your description of the schemes today useful though,
although not everyone will be aware of the reasons
why you have made them.

Maybe we can focus on base systems with as few
requirements for change as possible, since the
optimizations are largely designed to work with
existing systems.

They may then be able to slot into a base DNA
system without further significant modification
(depending on the spec).

> Another thanks for good work.  I am glad that MD BoF is moving 
> forward. 

Maybe with a slightly different focus, but I think
that DNA has good potential to provide solutions which
will work for MD.

Greg