Re: IP QoS issues
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IP QoS issues



> > On the Internet, where certain amount (50~30%) of bandwidth must
> > be reserved for best effort traffic, average queue length of QoS
> > assured traffic will be 2 or 3 that we don't have to worry about
> > jitter.

On today's Internet, I'd say that 100% of the traffic is 'reserved'
for best-effort traffic since there's little to no deployment of
QoS, class-of-service or differentiated services in the production
"backbone" of the Internet today.

> > For the time being, a point-to-point giga ether will be the
> > major link technology to offer QoS.
> 
> >From what I know Gigabit Ethernet has not QoS built in. I can't see how it
> can offer QoS with its inherited limitations. 802.3 is not designed with QoS
> in mind.

Well I could make the same statement about HSSI or SONET framing; the
trickyness is in the queueing disciplines, not the physical link layer.

> Depends always on the size of the ATM switch. Most switches that are used on
> the backbone (eg. Newbridge 36190 Core Services Switch) goes up to a maximum
> 2.000.000 VPCs and VCCs per switch. Generally, high-capacity ATM switches
> have adequate capacity.

First, it's extraordinarily unlikely that you'll see individual ATM VC's,
one per "QoS Flow" (whatever that means to you) because of the inherit
unscalability of that solution on "the backbone."  I don't know where
you got the notion that a Newbridge 36190 is representative of what
gets used on the Internet "backbone" today; I know for UUNET and other
large backbone operators, this isn't the case.  

However, the number of VC's isn't the problem.  This is the same misconception
that people have regarding the size of the routing table causing
the next Internet apocolypse; it not the SIZE but the CHURN RATE that will
kill you first.  The thing to worry about is the call setup rate in
the ATM switches, not the number of VCs it supports.

I now await the inevitable flood of bounced delivery notifications :-)  Should
be more interesting than reading about the IANA cabal.

Louis A. Mamakos                                    louie at uu.net,  uunet!louie
Senior Strategic Technology Manager
UUNET Technologies, Inc.                            Voice: +1 703 206 5823
3060 Williams Drive                                 Fax:   +1 703 206 5601
Fairfax, VA    22031-4648




Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.

Note: Messages sent to this list are the opinions of the senders and do not imply endorsement by the IETF.