Bytes Vs Packets - Random Early Detection (RED)

Hi,

Is RED in packet mode better than RED in byte mode ?
Which is good ?

Thx in advans,
Karthik Balaguru

"karthikbg" <karthik.balaguru@lntinfotech.com> hath wroth:

>Is RED in packet mode better than RED in byte mode ?
>Which is good ?
>
>Thx in advans,
>Karthik Balaguru

<http://en.wikipedia.org/wiki/Random_early_detection>

Good for what? UDP(non-adaptive) or TCP(adaptive) or mixed traffic?
Large or small packets? ATM or IP layer? With or without QoS (may be
impossible)? Using whos algorithm?

For what are you trying to optimize? Bandwidth or latency (for VoIP)?
This is far to general a question.

Hmmm... sounds like a student question. I think you need to do some
more reading before you ask others to do your homework. If not a
student, you need to explain what problem you are trying to solve. The
best method used to drop packets varies with just about everything
(type of traffic, protocols, buffer type, router horsepower, QoS,
etc).

Check the periodicals and publications from the IEEE and ACM at:
<http://scholar.google.com>

You're also asking in the wrong newsgroup. Try:
<news:comp.protocols.tcp-ip>
instead.

I found one old paper that might be relevant:
<http://www-nrg.ee.lbl.gov/floyd/REDaveraging.txt >


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On Feb 10, 9:26 pm, Jeff Liebermann <j...@comix.santa-cruz.ca.us>
wrote:
> "karthikbg" <karthik.balag...@lntinfotech.com> hath wroth:
>
> >Is RED in packet mode better than RED in byte mode ?
> >Which is good ?
>
> >Thx in advans,
> >Karthik Balaguru
>
> <http://en.wikipedia.org/wiki/Random_early_detection>
>
> Good for what? UDP(non-adaptive) or TCP(adaptive) or mixed traffic?
> Large or small packets? ATM or IP layer? With or without QoS (may be
> impossible)? Using whos algorithm?
>
> For what are you trying to optimize? Bandwidth or latency (for VoIP)?
> This is far to general a question.
>
> Hmmm... sounds like a student question. I think you need to do some
> more reading before you ask others to do your homework. If not a
> student, you need to explain what problem you are trying to solve. The
> best method used to drop packets varies with just about everything
> (type of traffic, protocols, buffer type, router horsepower, QoS,
> etc).
>
> Check the periodicals and publications from the IEEE and ACM at:
> <http://scholar.google.com>
>
> You're also asking in the wrong newsgroup. Try:
> <news:comp.protocols.tcp-ip>
> instead.
>
> I found one old paper that might be relevant:
> <http://www-nrg.ee.lbl.gov/floyd/REDaveraging.txt >
>
> --
> Jeff Liebermann j...@comix.santa-cruz.ca.us
> 150 Felker St #D http://www.LearnByDestroying.com
> Santa Cruz CA 95060http://802.11junk.com
> Skype: JeffLiebermann AE6KS 831-336-2558

Got the answer via google.

Byte's Advantage
=============
For a router where the transmission delay for a packet is largely a
function of the size of the packet in bytes, then measuring the queue
in bytes has the advantage that the average queue size corresponds to
the average queueing delay for a packet, in seconds.

Packet's Advantage
===============
For a router where the transmission delay for a packet is fixed,
regardless of the size of the packet in bytes, then measuring the
queue in packets gives the most accurate indication of the average
queueing delay for an arriving packet.

Thx,
Karthik Balaguru