Title: Using%20Partial%20Differential%20Equations%20to%20Modek%20TCP%20Mice%20and%20Elephantsin%20large%20IP%20Networks
1Using Partial Differential Equations to Modek TCP
Mice and Elephantsin large IP Networks
- M. Ajmone Marsan, M. Garetto, P. Giaccone,
- E. Leonardi, E.Schiattarella, A. Tarello
- Politecnico di Torino - Italy
Hong-Kong March 7-11 , 2004 TANGO
2Outline
- Dimensioning IP networks
- Queuing network models
- Fluid approaches
- Conclusions
3Consideration
- Over 90 of all Internet traffic is due to TCP
connections - TCP drives both the network behavior and the
performance perceived by end-users - Analytical models of TCP are a must for IP
network design and planning
4Consideration
- Accurate TCP models must consider
- closed loop behavior
- short-lived flows
- multi-bottleneck topologies
- AQM schemes (or droptail)
- QoS approaches, two-way traffic, ...
5Problem statement
2
finite flows (mice)
1
greedy flows
? URLs/sec
IP core
? URLs/sec
finite flows
3
N
greedy flows (elephants)
4
...
6Problem statement
Input variables only primitive network
parameters
- IP network channel data rates, node distances,
buffer sizes, AQM algorithms (or droptail), ... - TCP number of elephants, mice establishment
rates and file length distribution, segment size,
max window size, ...
Output variables
- IP network link utilizations, queuing delays,
packet loss probabilities, ... - TCP average elephant window size and
throughput, average mice completion times, ...
7Modeling approach
- Abandon a microscopic view of the IP network
behavior, and model packet flows and other system
parameters as fluids - The system is described with differential
equations - Solutions are obtained numerically
8Modeling approach
- A simple example
- One bottleneck link
- RED buffer
- Elephants only (no slow start)
9TCP model
dWs(t)/dt 1/Rs(t) Ws(t) ?s(t) / 2
- Where
- Ws(t) is the average window
- Rs(t) is the average round trip time
- ?s(t) is the congestion indication rate
- of TCP sources of class s at time t
10IP network model
dQk(t)/dt Ss Ws(t) (1-P(t)) / Rs(t) - C
1Qk(t)gt0
- Where
- Qk(t) is the length of queue k at time t
11IP network model
Rs(t) PDs Qk(t)/C
- Where
- PDs is the propagation delay for source s
12Problems
Difficult to deal with mice since the start time
of each mouse detemines the window dynamics over
time. One equation shoud be written for each
mouse Difficult to consider droptail buffers due
to the intrinsic burstiness of the loss process
experienced by sources
13Problems
Difficult to deal with mice since the start time
of each mouse detemines the window dynamics over
time. One equation shoud be written for each
mouse
14Our Approach
- Consider a population of TCP sources P(w,t) is
the number of TCP flows that at time t have
window not greater than w.
. -
Partial differential equations are obtained
15Basic source model
Where
16Mice Source Equations
17Fluid models extensions
- Slow start (mice)
- Finite window
- Threshold
- Fast recovery
- Droptail buffers
- Core network topologies
18Fluid models results
19Fluid models model results
20Fluid models NS results
21Fluid models model results
22Fluid models NS results
23Fluid models results
24Fluid models results
25Fluid models results
26Fluid models results
27Fluid models results
28Fluid models results
We obtained results for the GARR network with
over one million TCP flows, and link capacities
up to 2.5 Gb/s.
29Conclusions
- Fluid models today seem the most promising
approach to study large IP networks - Tools for the model development and solution are
sought - Efficient numerical techniques are a challenge
30Conclusions
- The modeling paradigms to study the Internet
behaviour are changing - This is surely due to scaling needs, but
probably also corresponds to a new phase of
maturity in Internet modeling - Reliable predictions of the behavior of
significant portions of the Internet are within
our reach
31Thank You !
32Outline
- The Internet today
- Dimensioning IP networks
- Queuing network models
- Fluid approaches
- Conclusions
33Source Internet Software Consortium
(http//www.isc.org/)
34Source Internet Traffic Report
(http//www.internettrafficreport.com/)
35Source Internet Traffic Report
(http//www.internettrafficreport.com/)
36Source Sprint ATL (http//ipmon.sprint.com/packst
at) April 7th 2003, 2.5 Gbps link
37Source Sprint ATL (http//ipmon.sprint.com/packst
at) April 7th 2003, 2.5 Gbps link
38Source Sprint ATL (http//ipmon.sprint.com/packst
at) April 7th 2003, 2.5 Gbps link
39Source Sprint ATL (http//ipmon.sprint.com/packst
at) April 7th 2003, 2.5 Gbps link
40Source Sprint ATL (http//ipmon.sprint.com/packst
at) April 7th 2003, 2.5 Gbps link
41And still growing ...
Subject news Internet still growing 70 to 150
per cent per year Date Mon, 23 Jun 2003 095545
-0400 (EDT) From CAnet-NEWS_at_canarie.ca ... Andre
w Odlyzko, director of the Digital Technology
Center at the University of Minnesota, ... says
Internet traffic is steadily growing about 70
percent to 150 percent per year. On a conference
call yesterday to discuss the results, he said
traffic growth slowed moderately over the last
couple of years, but it had mostly remained
constant for the past five years. ...
42Literature
V. Misra, W. Gong, D. Towsley, "Stochastic
Differential Equation Modeling and Analysis of
TCP Windowsize Behavior, Performance'99 T.
Bonald, "Comparison of TCP Reno and TCP Vegas via
Fluid Approximation," INRIA report no. 3563,
November 1998 V. Misra, W. Gong, D. Towsley, "A
Fluid-based Analysis of a Network of AQM Routers
Supporting TCP Flows with an Application to RED,
SIGCOMM 2000
43Literature
Y.Liu, F.Lo Presti, V.Misra, D.Towsley, Y.Gu,
"Fluid Models and Solutions for Large-Scale IP
Networks", SIGMETRICS 2003 F. Baccelli, D.Hong,
"Interaction of TCP flows as Billiards, Infocom
2003 F.Baccelli, D.Hong, "Flow Level Simulation
of Large IP Networks, Infocom 2003
44Literature
T. Lakshman and U. Madhow, "The performance of
TCP/IP for networks with high bandwidth-delay
products and random loss," IEEE/ACM Transactions
on Networking, vol. 5, no. 3, 1997. M.Ajmone
Marsan, E.de Souza e Silva, R.Lo Cigno, M.Meo,
An Approximate Markovian Model for TCP over
ATM, UKPEW '97 J. Padhye, V. Firoiu, D.
Towsley, J. Kurose, "A Stochastic Model of TCP
Reno Congestion Avoidance and Control, UMASS
CMPSCI Technical Report, Feb 1999.
45Literature
C.Casetti, M.Meo, A New Approach to Model the
Stationary Behavior of TCP Connections, Infocom
2000 M.Garetto, R.Lo Cigno, M.Meo, E.Alessio,
M.Ajmone Marsan, Modeling Short-Lived TCP
Connections with Open Multiclass Queueing
Networks, PfHSN 2002 A.Goel, M.Mitzenmacher,
"Exact Sampling of TCP Window States", Infocom
2002
46Consideration
Developing accurate analytical models of the
behavior of TCP is difficult. A number of
approaches have been proposed, some based on
sophisticated modeling tools.
47Fluid models results