From: roy bustos (roybustos@xxxxxxxxxxx)
Date: Wed Feb 13 2002 - 14:18:02 GMT-3
Yes True but, WRED drops packets to decrease window size(TCP) to reduce
congestion,
so it doesn't relieve congestion for other protocols.
DOCS:
By randomly dropping packets prior to periods of high congestion, WRED tells
the packet source to decrease its transmission rate. If the packet source is
using TCP, it will decrease its transmission rate until all the packets
reach their destination, which indicates that the congestion is cleared.
WRED generally drops packets selectively based on IP precedence. Packets
with a higher IP precedence are less likely to be dropped than packets with
a lower precedence. Thus, the higher the priority of a packet, the higher
the probability that the packet will be delivered.
WRED reduces the chances of tail drop by selectively dropping packets when
the output interface begins to show signs of congestion. By dropping some
packets early rather than waiting until the queue is full, WRED avoids
dropping large numbers of packets at once and minimizes the chances of
global synchronization. Thus, WRED allows the transmission line to be used
fully at all times.
In addition, WRED statistically drops more packets from large users than
small. Therefore, traffic sources that generate the most traffic are more
likely to be slowed down than traffic sources that generate little traffic.
WRED avoids the globalization problems that occur when tail drop is used as
the congestion avoidance mechanism. Global synchronization manifests when
multiple TCP hosts reduce their transmission rates in response to packet
dropping, then increase their transmission rates once again when the
congestion is reduced.
WRED is only useful when the bulk of the traffic is TCP/IP traffic. With
TCP, dropped packets indicate congestion, so the packet source will reduce
its transmission rate. With other protocols, packet sources may not respond
or may resend dropped packets at the same rate. Thus, dropping packets does
not decrease congestion.
Roy
----- Original Message -----
From: <RSiddappa@NECBNS.com>
To: <brian@cyscoexpert.com>; <roybustos@hotmail.com>; <mamoor@ieee.org>
Cc: <ccielab@groupstudy.com>
Sent: Wednesday, February 13, 2002 7:45 AM
Subject: RE: CAR and WRED
> Brian,
>
> I agree. I was little short of sleep when I wrote that, Sorry.
>
> R.
>
>
> -----Original Message-----
> From: Brian McGahan [mailto:brian@cyscoexpert.com]
> Sent: Wednesday, February 13, 2002 9:44 AM
> To: Siddappa, Rajeev; roybustos@hotmail.com; mamoor@ieee.org
> Cc: ccielab@groupstudy.com
> Subject: Re: CAR and WRED
>
>
> You're right in the statement that untagged IP traffic, and non-IP traffic
> will be treated as IP Prec 0, but not that they will be dropped. The
lower
> the precedence value, the higher the probability of being dropped. This
is
> not to say that ANY traffic will be dropped, just that IP prec 7 has less
> probability of being dropped than IP prec 0.
>
>
> Brian McGahan
> CCIE #8593
> brian@cyscoexpert.com
>
> CyscoExpert Corporation
> Internetwork Consulting & Training
> http://www.cyscoexpert.com
> Voice: 847.674.3392
> Fax: 847.674.2625
>
> ----- Original Message -----
> From: <RSiddappa@NECBNS.com>
> To: <roybustos@hotmail.com>; <mamoor@ieee.org>
> Cc: <ccielab@groupstudy.com>
> Sent: Wednesday, February 13, 2002 1:50 AM
> Subject: RE: CAR and WRED
>
>
> > Since the WRED is based on eh precedence value in the TOS field of the
> IP,
> > it good for TCP and any other non IP traffic will be treated with
> precedence
> > of 0 and will be dropped.
> >
> > My 2 cents,
> >
> > R.
> >
> >
> >
> >
> >
> >
> >
> >
> > -----Original Message-----
> > From: roy bustos [mailto:roybustos@hotmail.com]
> > Sent: Tuesday, February 12, 2002 3:06 PM
> > To: Ahmed Mamoor Amimi
> > Cc: ccielab@groupstudy.com
> > Subject: Re: CAR and WRED
> >
> >
> > WRED only works with TCP and is based on congestion, it will randomly
> drop
> > packets once your load reaches congestion,
> > CAR is a form of rate limiting, and will perform even without
congestion.
> >
> > Roy
> > ----- Original Message -----
> > From: "Ahmed Mamoor Amimi" <mamoor@ieee.org>
> > To: <ccielab@groupstudy.com>
> > Sent: Tuesday, February 12, 2002 11:41 AM
> > Subject: CAR and WRED
> >
> >
> > > my thinking about CAR and WRED :
> > >
> > > CAR
> > > By CAR u can set normal and max burst then tell if the rate drop from
> > > normal then apply precedence or drop or ATM-CLP or QoS or trasmit.
> > > General assumption:
> > > set min xxx
> > > max xxx
> > > if min<flow of packet
> > > then
> > > drop or clp or prec or qos or transmit
> > > end
> > >
> > >
> > > WRED
> > > Use the random-detect command to enable WRED, which randomly discards
> > > packets during congestion based on IP precedence settings
> > > General assumption:
> > > if RED<flow of packet
> > > then
> > > {(set precedence 0-7) and (threshold 1-4096)}
> > > or
> > > set precedence to rsvp packets
> > >
> > >
> > >
> > > my thinking is that they both are some what same but CAR is more
> flexible
> > as
> > > it gives u the upper and lower bond of limit
> > > to traffic.
> > >
> > >
> > > -Mamoor
This archive was generated by hypermail 2.1.4 : Thu Jun 20 2002 - 13:46:21 GMT-3