From: Gary Duncanson (gary.duncanson@googlemail.com)
Date: Sun Jul 22 2007 - 14:19:02 ART
Good points Paul.
Gary
----- Original Message -----
From: "Paul Dardinski" <pauld@marshallcomm.com>
To: <johngibson1541@yahoo.com>; <ccielab@groupstudy.com>
Sent: Sunday, July 22, 2007 11:39 AM
Subject: RE: We want to prevent packet lost? But TCP needs packet lost.
> John,
>
> QoS is not designed to protect against packet loss. It's only purpose is
> to ensure preferential treatment of one packet type over another in the
> area of latency, jitter or in the decision of which packet to drop in
> the case of data overflow.
>
> Data overflow is not a generally a function of a well designed router or
> switch. QoS is not there to repair an overutilized CPU. Data overflow is
> most often derived from temporary oversubscription of a link, hence the
> need for QoS. If a link if regularly oversubscribed, then all the QoS in
> the world won't fix it, the pipe has to be made larger.
>
> Network hardware is designed for certain throughput capacities defined
> by the manufacturer. Generally they are designed with the capacities of
> the possible ports allowed in the box as much as possible (ie. switches
> should be nonblocking across the backplane up to the capacity of traffic
> they are rated for). The core CPU as you had mentioned often handles the
> backplane throughput. Most enterprise equipment today uses distributed
> architectures allowing multiple CPU's across the architecture, each with
> defined tasks to ensure that processing and serialization are minimized
> as much as possible. QoS queuing purpose is to have a place to hold
> traffic for differentiation and availability, not to compensate for the
> CPU.
>
> PD (#16842
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> johngibson1541@yahoo.com
> Sent: Saturday, July 21, 2007 11:25 PM
> To: ccielab@groupstudy.com
> Subject: Re: We want to prevent packet lost? But TCP needs packet lost.
>
> Even smarter these people. The TCP protocol is designed to
> have packet lost. And they can just come up with more and more
> sophisticated queues every month to "prevent packet lost".
>
> It is designed to have packet lost, and they come to their
> work every day to "prevent packet lost". How smart is that.
>
> They will never need to create any thing and they have
> a job guaranteed for life.
>
> This is like the lawyers working with law makers to come up
> with more sophisticated forms to do the same thing every
> month so that citizens need lawyers.
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Sat Aug 18 2007 - 08:17:41 ART