Re: [Sipping] FW:WGLC Review: draft-ietf-sipping-presence-scaling-requirements-01.txt

["Schmidt, Christian 1. (NSN - DE/Munich)" <christian.1.schmidt at nsn.com>]

Hi Paul,

Sorry for the late reply, but the harddisk of my notebook has refused
further service this week.

I like your new version of REQ-007, it covers the point I raises.

A related question: What about Denial of Service attacks? Should we
provide requirements, to restrict the risk of DoS attacks? For example
to restrict the number of watchers per presentity or to restrict the
number of changes in presence status per time-unit?

Regards
Christian 



 

-----Original Message-----
From: ext Paul Kyzivat [mailto:pkyzivat at cisco.com] 
Sent: Freitag, 19. September 2008 15:31
To: Schmidt, Christian 1. (NSN - DE/Munich)
Cc: ext Gonzalo Camarillo; sipping at ietf.org;
draft-ietf-sipping-presence-scaling-requirements at tools.ietf.org; Mary
Barnes
Subject: Re: [Sipping] FW:WGLC Review:
draft-ietf-sipping-presence-scaling-requirements-01.txt



Schmidt, Christian 1. (NSN - DE/Munich) wrote:
> Hi,
> 
> A short comment / queston, concerning a scalility requirement:
> 
>    o  REQ-007: Presence systems (intra or inter-domain) SHOULD scale
in
>       linear proportion to the number of watchers and presentities in
>       the system.
> 
> This requirement takes not into account the relationsship between
presentities
> And watchers. Meaning for example in adding few presentities with a
huge amount of watchers,
> This linear proportion can not be achieved. Therefore I think, this
has to
> Be taken into account in this requirement.

>    o  REQ-007: Presence systems (intra or inter-domain) SHOULD scale
in
>       linear proportion to the number of watchers and presentities in
>       the system. Precondition ist that the number of watchers of a
presentity and the
>       number of precentities a watcher can monitor is restricted.

How about:

    o  REQ-007: Presence systems (intra or inter-domain) SHOULD scale in
       linear proportion to the number of watchers and presentities in
       the system, so long as the average number of watchers per
       presentity has a fixed bound.

Namely, its not necessary to bound any particular user's behavior so 
long as the average is fixed rather than proportional to the number of 
presentities.

Over the short term that is likely to be true by accident. Over longer 
periods of time, as the system becomes ubiquitous, the average number is

likely to rise but maybe not in proportion to the number of users in the

system.

This behavior could probably be constrained if there was a cost to the 
end user associated with having a buddy.

	Thanks,
	Paul
_______________________________________________
Sipping mailing list  https://www.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors at cs.columbia.edu for questions on current sip
Use sip at ietf.org for new developments of core SIP