[e2e] congestion correlation?
Andrew Odlyzko
amo at research.att.com
Sat May 5 19:04:16 PDT 2001
As Vern Paxson noted already, he has collected extensive evidence
that corrections are weak, if they exist at all. This should not
be too surprising, since, among other things, "hot potato routing"
means that packets often traverse almost disjoint paths. Futher,
even when they do travel over the same path, there is often
substantial asymmetry in the traffic load in the two directions. This
is true even within the US. but very noticeable on trans-Atlantic
and trans-Pacific connections. As an example, if you look at the
data for JANET, the British academic and research network (available
at <http://bill.ja.net/>) you will find the following ratio for
(traffic from the US to Britain) / (Britain to US traffic):
month ratio
-------- ----
March 97 1.26
March 98 1.98
March 99 2.05
March 00 3.27
March 01 2.70
(An increase in this ratio also held true for some other links in
the late 1990s, and I am watching with interest whether it will
reverse course and shrink, as the ratio for JANET may be doing.)
Capacities are generally completely symmetric. This is true
throughout the telecom world (except for newer technologies
at the edges, like DSL, cable modems, and satellite data), a
legacy of the voice telephony world. (Frame Relay offers
customers the option of asking for asymmetric PVCs, but this
is just how the capacity is sold, as the Frame links themselves
are symmetric.) In my Globecom '99 paper I used this as an
argument against QoS: Basically the reasoning was that with
far less effort than is going into developing and deploying
QoS, one could develop a transport infrastructure that would
take advantage of the asymmetries in traffic patterns, which
tend to persist for extended periods of time, and that this
would provide at least as much of a gain in performance.
Andrew Odlyzko
-----Original message-----
From: Adam Wolisz <wolisz at ee.tu-berlin.de>
To: Sam Liang <sliang at DSG.Stanford.EDU>, end2end-interest at postel.org
Subject: Re: [e2e] congestion correlation?
Date: Thu, 03 May 2001 10:37:57 +0200
The answer is: NO
For example in 1999 we have done measurements both ways
between Berlin, Germany and Berkeley, CA:
Losses in the direction; Berkeley=>Berlin
have been much higher.
We believe that there is much more content being
transported from US to Europe than another way round.
And the capacities are not asymmetric enough.
(if at all).
Btw: Does anybody know if there is at all an asymmetry
of the resources between US and Europe?
You could contact Dorgham Sisalem,
sisalem at fokus.gmd.de
for more details on this measurements, done in the
context of his Ph.D Thesis on TCP friendly
UDP traffic, available from
ftp://ftp.fokus.gmd.de/pub/glone/sa.da/Sisa0010:TCP.pdf
Adam Wolisz
Sam Liang wrote:
> Hello,
>
> Is there any study done on whether there is any congestion correlation
> on the two directions between two endpoints A and B? Suppose A is sending
> data to B. When there is congestion and packet loss from A to B, do we
> expect packet loss from B to A at the same time?
>
> Thanks,
>
> Sam
************************************************************************
Andrew Odlyzko amo at research.att.com
AT&T Labs - Research voice: 973-360-8410
http://www.research.att.com/~amo fax: 973-360-8178
************************************************************************
More information about the end2end-interest
mailing list