CMPE 150 Winter 2009

Transcription

1 CMPE 150 Winter 2009 Lecture 17 March 5, 2009 P.E. Mantey

2 CMPE Introduction to Computer Networks Instructor: Patrick Mantey / t / Office: Engr. 2 Room 595J Office hours: Tues 3-5 PM, Mon 5-6 PM* TA: Anselm Kia akia@soe.ucsc.edu Web site: Text: Tannenbaum: Computer Networks (4 th edition available in bookstore, etc. )

3 Syllabus

4 Text Readings Today: Chapter 6, Section (TCP Performance) Tuesday March 10 Chapter 7.1 (DNS), Chapter 7.4 (Multimedia)

5 Internet Layering Level 5 -- Application Layer (rlogin, ftp, SMTP, POP3, IMAP, HTTP..) Level 4 -- Transport Layer(a.k.a a Host-to-Host) to Host) (TCP, UDP) Level 3 -- Network Layer (a.k.a. Internet) (IP, ICMP, ARP) Level 2 -- (Data) Link Layer / MAC sub-layer (a.k.a. Network Interface or Network Access Layer) Level 1 -- Physical Layer

6 Transport Layer Connection Oriented Service Addressing / Access Points Connection Request / Establishment Delayed Duplicate Packet Problem Timer management Connection points Connection Release

7 Today s Agenda Transport Layer TCP: Timers / RTT estimates Exponential backoff Wireless vs. wired Transactional TCP Performance

8

9 Port Detective

10

11

12 3389 Remote desktop (Windows XP) Reference: p g p p_ p ========================================================== See also:

13 TCP Congestion Control (a) A fast network feeding a low capacity receiver.

14 Jacobson s Algorithm 1 Determining the round-trip time: TCP keeps RTT variable. When segment sent, TCP measures how long it takes to get ACK back (M). RTT = alpha*rtt + (1-alpha)M. alpha: smoothing factor; determines weight given to previous estimate. Typically, alpha=7/8.

15 Jacobson s Algorithm 2 Determining timeout value: Measure RTT variation, or RTT-M. Keeps smoothed value of cumulative variation D=alpha*D+(1-alpha) RTT-M. Alpha may or may not be the same as value used to smooth RTT. Timeout = RTT+4*D.

16 Karn s Algorithm How to account for ACKs of retransmitted segments? Count it for first or second transmission? Karn proposed not to update RTT on any retransmitted segment. Instead RTT is doubled on each failure until segments get through.

17 TCP Congestion Control: Example cwin timeout threshold threshold time

18 Persistence Timer Prevents deadlock if an window update packet is lost and advertised window = 0. When persistence timer goes off, sender probes receiver; receiver replies with its current advertised window. If = 0, persistence timer is set again.

19 Keepalive Timer Goes off when a connection is idle for a long time. Causes one side to check whether the other side is still alive. If no answer, connection terminated.

20 TCP Timer Management (a) Probability density of ACK arrival times in the data link layer. (b) Probability density of ACK arrival times for TCP.

21 TCP Retransmission Timer When segment sent, retransmission timer starts. If segment ACKed, timer stops. If time out, segment retransmitted and timer starts eou,seg e e s ed d e s s again.

22 How to set timer? Based on round-trip time: time between a segment is sent and ACK comes back. If timer is too short, unnecessary retransmissions. If timer is too long, long retransmission delay.

23 Jacobson s Algorithm 1 Determining the round-trip time: TCP keeps RTT variable. When segment sent, TCP measures how long it takes to get ACK back (M). RTT = alpha*rtt + (1-alpha)M. alpha: smoothing factor; determines weight given to previous estimate. Typically, alpha=7/8.

24 Jacobson s Algorithm 2 Determining timeout value: Measure RTT variation, or RTT-M. Keeps smoothed value of cumulative variation D=alpha*D+(1-alpha) RTT-M. Alpha may or may not be the same as value used to smooth RTT. Timeout = RTT+4*D.

25 Karn s Algorithm How to account for ACKs of retransmitted segments? Count it for first or second transmission? Karn proposed not to update RTT on any retransmitted segment. Instead RTT is doubled on each failure until segments get through.

26 Persistence Timer Prevents deadlock if an window update packet is lost and advertised window = 0. When persistence timer goes off, sender probes receiver; receiver replies with its current advertised window. If = 0, persistence timer is set again.

27 Keepalive Timer Goes off when a connection is idle for a long time. Causes one side to check whether the other side is still alive. If no answer, connection terminated.

28 Performance Issues Performance Problems in Computer Networks Network Performance Measurement System Design for Better Performance Fast TPDU Processing Protocols for Gigabit Networks

29

30

31

32 System Design for Better Performance Rules: CPU speed is more important than network speed. Reduce packet count to reduce software overhead. Minimize context switches. Minimize copying. You can buy more bandwidth but not lower delay. Avoiding congestion is better than recovering from it. Avoid timeouts.

33 Wireless TCP and UDP Splitting a TCP connection into two connections.

34 Wireless TCP 1 According to layered system design principles, transport protocol should be independent of underlying technology. However, wireless networks invalidate this principle. Ignoring properties of wireless medium can lead to poor TCP performance. Problem: TCP s congestion control.

35 Wireless TCP 2 Problem: packet loss as congestion indicator. When retransmission timer times out, sender slows down. Wireless links are lossy! Dealing with losses in this case should be re-sending lost segments asap.

36 Indirect TCP (I-TCP) [Bakne and Badrinath, 1995]. Split TCP connection in 2: one from sender to base station and the other from base station to receiver. Base station serves as repeater : copies segments between connections in both directions. Connections are homogeneous; timeouts on 1st. connection, slow down sender. Problem: violates TCP s e2e ness. Example: ACKs to sender mean base station received segments, not necessarily receiver.

37 Snoop TCP [Balakrishnan et al., 1995]. Does not break connection. Modifications to base station s network layer code. Snooping agent on base station observes and caches TCP segments sent to mobile host and ACKs coming back. If it doesn t see an ACK for a segment or sees duplicate ACKs, it times out and retransmits. But source may time out anyway.