From: Mark Miller (markmiller@alltel.net)
Date: Mon Apr 21 2003 - 23:25:44 GMT-3
This is taken from page 489 of ciscopress Integrating Voice and
Data Networks.
Consider for example, that you use RSVP or IP Precedence in
conjunction with VoIP. Using RED on such traffic flows is pointless.
Actually, it is worse than pointless - it is counterproductive. You
can effectively disable WRED for such flows, while still preserving
WRED for other flow types, with the commands found in the following
example.
r2(config-if)#random-detect precedence 5 100 101 65536
r2(config-if)#random-detect rsvp 100 101 65536
This sets the drop probability at 0.0015 percent.
It goes on to explain that if you do not have memory resource
constraints, you can increase the queue depths as high as 4096, which
pretty much sets the chances of these specified flows being dropped at 0%.
Hope this helps,
Mark
----- Original Message -----
From: "Ray Stevens" <cisco-guy@rogers.com>
To: <ccielab@groupstudy.com>
Sent: Monday, April 21, 2003 7:18 PM
Subject: RE: [RE]Re: [RE]random-detect precedence 5 40 60 20 (OUT OF OFFICE)
> This is how I understand it as well. but when reading it indicates that
> WRED will drop all packets simulating tail-drop when the queue is full.
Is
> this refering to when the max-threshold is reashed or not. depending on
the
> sources I have read sometimes it sounds like the probability denomiintor
is
> the number of packets dropped when it reaches the min and incrases to
> dropping all when at the maximum-threshold, and others say that it will
> start dropping a percentage of the marked probabiltiy at minimum nd
increase
> to full marked proability when at max-threshold.
>
> if the second is correct at what point would WRED start dropping all
> packets.
>
>
> exerpt from ISO Quality Of Service book.
>
> WRED is usually used in the core routers of a network, rather than at the
> edge. WRED uses these
> precedences to determine how it treats different types of traffic.
> When a packet arrives, the following events occur:
> 1. The average queue size is calculated.
> 2. If the average is less than the minimum queue threshold, the arriving
> packet is queued.
> 3. If the average is between the minimum queue threshold for that type of
> traffic and the maximum
> threshold for the interface, the packet is either dropped or queued,
> depending on the packet drop
> probability for that type of traffic.
> 4. If the average queue size is greater than the maximum threshold, the
> packet is dropped.
>
> This would indicate once it goes over the max treshold it drops all which
> would seem like a step increase from say dropping 1 out of 50 at max
> threshold and dropping all once it goes over.
>
>
> Ray Stevens
> CCNP, CCDP, SCSA
>
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> midaskim
> Sent: Friday, April 18, 2003 4:03 AM
> To: dhuskey@bpop.com
> Cc: ccielab@groupstudy.com
> Subject: [RE]Re: [RE]random-detect precedence 5 40 60 20 (OUT OF OFFICE)
>
>
> Thanks..
>
> But I found the answer for my question.
>
> The packet drop probability is based on the minimum threshold, maximum
> threshold, and mark probability denominator.
>
> When the average queue depth is above the minimum threshold, RED starts
> dropping packets. The rate of packet drop increases linearly as the
> average queue size increases until the average queue size reaches the
> maximum threshold.
>
> The mark probability denominator is the fraction of packets dropped when
> the average queue depth is at the maximum threshold.
>
> ---------[ 9^@: 8^@O 3;?k ]----------
> A&8q : Re: [RE]random-detect precedence 5 40 60 20 (OUT OF OFFICE)
> 3/B% : Fri, 18 Apr 2003 02:09:44 -0500
> :83=@L : "David Huskey" <dhuskey@bpop.com>
> 9^4B@L : <janggulove@hanmail.net>
>
> I will be out of the office Friday April 18th
> Returning on Monday April 21st.
>
> If you need immediate attention please call
> THE HELP DESK @ 847-994-5890
>
> [IMAGE]
>
> [IMAGE]
>
> [IMAGE]
>
> midaskim
>
> [IMAGE]
>
> [IMAGE]
>
> "?l8. @NEM3], Daum" http://www.daum.net !:Fr;}>24B 9+7a GQ8^@O3]!;
>
> [IMAGE]
>
> GA8.9L>v 8^@O =a:8<L>n?d?
> 60!Av =:E2+?k7.H.@e+%a 3*88@G 8BCc 8^@O!
>
> [IMAGE]
>
> [IMAGE]
>
> GQ8^@O3] =:FTA&7N =C=:E[
> =:FT >x4B FmAvGT! GQ8^@O3] FmAvGT!
>
> [IMAGE]
This archive was generated by hypermail 2.1.4 : Thu May 01 2003 - 13:36:00 GMT-3