Modeling TCP Latency

1
Modeling TCP Latency

• Cardwell,Savage, and Anderson
• INFOCOM2000

2
Abstract
3
I. Introduction

• Many of todays popular Internet applications,
  including WWW, e-mail, FTP, Usenet news, remote
  login, use TCP as a transport protocols.
• Analytical models proposed in terms of round trip
  time and packet loss rate
• Achieving better understanding of the sensitivity
  of TCP performance to network parameters
• Helping the design of active queue management
  schemes and TCP-friendly multicast protocols

4

• The analytical models proposed to date can be
  split into two broad classes
• Models for steady-state bulk transfer throughput
  that suffers at least one loss
• But, the majority of TCP flows traveling the
  wide-area Internet are very short with mean sizes
  around 10KB.
• They often spend their entire lifetime in the
  slow start mode, without suffering a single loss.
• Models for short flows that suffer no packet loss
• They Does not consider delayed
  acknowledgements, sender or receiver buffering
  limitations, alternative initial congestion
  windows, or losses during connection
  establishment, each of which can have a dramatic
  performance impact.

5

• This paper proposes a new model for TCP
  performance that integrates the results from both
  classes of models.

6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
(No Transcript)