Re: [EFM] Network timing, ATM, ADSL/VDSL and EFM
On Fri, 28 Sep 2001 17:30:01 -0700 "Andrew Smith" wrote:
> FWIW, I've always used the rule of thumb that for N bits-worth of QoS
> complexity you build into a network node, you get to use another 1/2^^(N+1)
> fraction of the potential capacity, or something like that, for small values
> of N. Of course this is a very rough rule and applies only to class-based
> queueing nodes, not individual flows. It also doesn't take account of the
> complexity of any mechanisms to police traffic as it enters the network.
> Note also that this is layer-independent. So, for example:
>
> Bits of QoS Usable cap. Example
> 0 50% Switched FullDup Ethernet
> 1 75% 3Com's PACE(tm)
> 3 94% 802.1D traffic classes
> 6 99% IETF's Diffserv
>
> But the point is that it is a game of rapidly dimishing returns: people
> shouldn't knock the idea of merely "throwing bandwidth at the problem" - in
> many economic situations that is exactly the *right* thing to do.
Andrew;
I agree with all of the above, but I would also like to add
the additional comment that this only addresses the theortical payload
capacity of the link, not the over all cost of service. For example,
as others have pointed out, among the larger problems with ATM was the
cost of the end equipment and the intangible cost of configuring a
network of complex devices. The latter cost is often overlooked,
except in practice when it becomes a nasty suprise.
There is an IP protocol designed to provide some measure of
QoS called Multi-protocol Label Switching (MPLS). I probably
shouldn't bring up MPLS as an example, because some love it and some
loath it, but I will point out that it sure as heck is expensive to
deploy in terms of equipment costs and the labor cost of maintaining
the additional complexity of configuration has a lot of bean-counters
waking up in the middle of the night with cold sweats. The cost of
MPLS is dependent on these two factors, not the additional overhead in
bits.
regards,
fletcher