[ippm] [e2e] Mathematical analysis of e2e lable switching path
alok.dube at apara.com
Fri Oct 4 12:33:01 PDT 2002
to add to his point,
i would really love a "document" which poses responses to the "jitter"
problem for MPLS...or benchmarks on the same..
thats the only thing that can limit MPLS from sweeping over circuit switch
as we all know, MPLS strives to achieve the performance of a circuit
switched network over a packet switched network...
----- Original Message -----
From: Jing Shen
To: Kavé Salamatian ; end2end-interest at postel.org
Cc: ippm at advanced.org
Sent: Friday, October 04, 2002 2:23 PM
Subject: RE: [ippm] [e2e] Mathematical analysis of e2e lable switching
Thanks for all those response.
To my limited knowledge on MPLS implementation, many Telecommunication
have adopted MPLS in their service network, such as UUnet, AT&T. In
China, ChinaTelecom has built up
a experimental MPLS backbone accross multiple provinces, and China Unicom
also built up a backbone alike.
In deed, it seems MPLS network is not put directly to commercial usage in
China but some of stuff in ChinaTelecom
told me that they do care the VPN capacity, TE capacity and high
performance promised by MPLS. The problem is
the operators seem could not catch up with the complex configuration so
fast. So, I don't understand
why MPLS will not be used in large scale networks?
To the performance difference between MPLS network and IP network, I
totally agree with you IP router have achieve
the same performance with MPLS. But, to our experiments with comparation
between LSP and IP routing , there do exist
some difference between LSP and IP routing. We've built a very small
testbed to compare difference between LSP& IP routing,
and we found that LSP shows lower jitter and packet loss rate under
resource reservation scheme and best-effort service under low
load. This is the reason why I asked the question.
Kav¨¦ Salamatian <salamat at rp.lip6.fr> µÄÕýÎÄ£º
first of all I should mention that MPLS deployment is far from being
don't think that it has been deployed even in a large network operator)
I largely suspect that it will never be deployed in a large scale to be
primary backbone technology !!!!
From the point of view of measurement, LSP networks should not have too
difference from traditional IP networks as nowadays IP routing
is getting very close to Label switching performance, meaning that the
delay of crossing an LSP network should not be significantly larger than
IP network. The difference may come from a simpler traffic engineering
MPLS. Meaning that we should evaluate the effect of traffic
performance, not the particuliar behaviour of MPLS network.
De?: ippm-admin at advance! d.org [mailto:ippm-admin at advanced.org]De la
Envoy¨¦?: dimanche 29 septembre 2002 04:28
¨¤?: end2end-interest at postel.org
Cc?: ippm at advanced.org
Objet?: [ippm] [e2e] Mathematical analysis of e2e lable switching path
I've some question on e2e performance analysis of
label switching path.
There has been many research on e2e performance
analysis for IP routing. Nearly all of them take the
model of "network of queue", and under assumpiton of
Possion Arrival, exponential flow size etc. To my
limited knowledge, this is not to the state of
Internet which experiences LRD. On the other hand,
different router archtecture must have different
effect on transmission performance.
As MPLS proceeds to be the primary backbone
technology, I think there must be something new
introduced into the e2e performance in internet.
So, I want to know, whether there is some work with
methematical! modeling of LSP networks? and, are there
anyone would do me a favor to recommend some refrence
book, web page, research paper or the like ?
Thank you very much.
State Key Lab of CAD&CG
HangZhou, ZheJiang Province 310027
Do You Yahoo!?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the end2end-interest