CMPE 80N Spring 2003 Week 9

1
CMPE 80N Spring 2003Week 9

• Introduction to Networks and the Internet

2
Announcements

• Quiz 3 on 06.03.
• Internet History video on Thu, 05.29.
• Project 2 due on 06.05 (by midnight).
• Look at the instructions on Project 2s Web page.
• Final exam on 06.09 from 730-1030pm.
• Comprehensive.
• Location BE 152 and BE 165 (overflow room).

3
Today

• TCP (contd).
• UDP.

4
Flow Control

• Flow control is necessary so that source doesnt
  transmit too fast for given receiver.
• E.g., a fast server trying to send 1Gb/s data to
  a small PC.
• Without some form of control, some data will get
  lost.

5
TCP Flow Control

• Sliding window.
• Receivers advertised window.
• Size of advertised window related to receivers
  buffer space.
• Sender can send data up to receivers advertised
  window.

6
TCP Sliding Window
7
TCP Flow Control Example
App. writes 2K of data
4K
2KSEQ0
2K
ACK2048 WIN2048
App. does 3K write
2K SEQ2048
0
Sender blocked
App. reads 2K of data
ACK4096 WIN0
ACK4096 WIN2048
2K
1K SEQ4096
Sender may send up to 2K
1K
8
Circuit- versus Packet Switching
9
Circuit Switching

• Old telephone technology.
• For each connection, switches establish physical
  circuit.

10
Circuit Switching - Example
Switch
Switching offices
11
Circuit Switching (contd)

• Switches are set up at the beginning of the
  connection and maintained throughout the
  connection.
• Network resources reserved and dedicated from
  sender to receiver.
• Not a very efficient strategy.
• A connection holds a physical line even during
  silence periods (when there is nothing to
  transmit)

12
Packet Switching

• Sharing by taking turns.
• Analogy conveyor belt in a warehouse.
• Items are picked from the storage room and placed
  on the conveyor belt every time a customer makes
  an order.
• Different customers may request a different
  number of items
• Different users items may be interspersed on the
  conveyor belt (they are multiplexed).
• Networks use a similar idea.
• Packet switching
• Packetize data to transfer.
• Multiplex it onto the wire.
• Packets from different connections share the same
  link.

13
Packet Switching Example
Payload
Header
A
C
D
B
14
Packet Switching

• Each packet is composed by payload (the data we
  want to transmit) and header.
• The header contains information useful for
  transmission, such as
• Source (senders) address.
• Destination (recipients) address.

15
Packet Switching (contd)

• The header introduces overhead, that is,
  additional bits to be sent.
• Tradeoff?
• What happens if the payload is too small?
• Or too big?

16
Circuit Switching vs Packet Switching

• Circuit switching
• Must set up a connection (initial delay)
• Resources are dedicated
• Therefore they may be used inefficiently!

• Packet switching
• Very small set-up delay.
• Efficient shared use of resources.
• But, may may result in too many packets being
  injected into the network.

17
Queuing
Packets are queued inside router, waiting to be
transmitted on outgoing link.
18
Queuing (contd)

• What happens if packets are coming in faster than
  router is able to forward them?
• Needs to queue the packets as they are waiting
  to be transmitted.
• What happens when a queue becomes too long?
• Packets are dropped!
• How can the receiver detected that a packet was
  dropped?
• Look at the packet sequence number!

19
Congestion

• Total offered load exceeds what network can
  handle.
• Congestion collapse
• When congestion occurs, packets get dropped.
• Due to packet loss, packets get retransmitted.
• Congestion gets worse and worse!
• Snow ball effect.

20
Congestion Control

• Why do it at the transport layer?
• Real fix to congestion is to slow down sender.
• Use law of conservation of packets.
• Keep number of packets in the network constant.
• Dont inject new packet until old one leaves.
• Congestion indicator packet loss.

21
TCP and Congestion Control

• Interprets packet loss as an indicator of
  congestion
• When it senses packet loss, it slows down the
  rate of packet transmission
• When packets are received correctly, sends
  packets faster
• Still within the limits of the sliding window

22
TCP Congestion Control

• Like, flow control, also window based.
• Sender keeps congestion window.
• Goal figure out size of congestion window.
• Trade-off?

23
TCP Congestion Control (contd)

• Slow start Jacobson 1988
• Connections congestion window starts at 1
  segment.
• If segment ACKed before time out, cwincwin1.
• As ACKs come in, current cwin is increased by 1.
• Exponential increase 1, 2, 4, 8, etc.

24
TCP Congestion Control (contd)

• Congestion Avoidance
• Slow start is aggressive exponential increase.
• Start low.
• But increase fast until finds available
  bandwidth.
• Then, cwin grows linearly.
• Upper bound set by flow control, I.e., maximum
  number of packets receiver can handle.

25
TCP Congestion Avoidance

• Congestion Avoidance
• cwin grows linearly until it reaches receivers
  advertised window.
• If timeout, thresholdcwin/2 and cwin1.
• Re-enters slow-start until cwinthreshold.

26
TCP Congestion Control Example
cwin
timeout
threshold
threshold
time
27
TCP Retransmission Timer

• When segment sent, retransmission timer starts.
• If segment ACKed, timer stops.
• If time out, segment retransmitted and timer
  starts again.

28
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.

29
TCP Segment Header
0
31
16
15
Source port
Destination port
Sequence number
Acknowledgment number
Header length
P
R
S
F
U
A
Window size
Checksum
Urgent pointer
Options (0 or more 32-bit words)
Data
30
TCP Header Fields

• Source and destination ports identify connection
  end points.
• Sequence number.
• Acknowledgment number specifies next byte
  expected.
• TCP header length how many 32-bit words are
  contained in header.
• 6-bit unused field.

31
TCP Header Fields (contd)

• 6 1-bit flags
• URG indicate urgent data present urgent
  pointer gives byte offset from current sequence
  number where urgent data is.
• ACK indicates whether segment contains
  acknowledgment if 0, acknowledgement number
  field ignored.
• PUSH indicates PUSHed data so receiver delivers
  it to application immediately.

32
TCP Header Fields (contd)

• Flags (contd)
• RST used to reset connection, reject invalid
  segment, or refuse to open connection.
• SYN used to establish connection connection
  request, SYN1, ACK0.
• FIN used to release connection.
• Window size how many bytes can be sent starting
  at acknowledgment number.

33
TCP Header Fields (contd)

• Checksum checksums the headerdatapseudo-header.
  
• Options provide way to add extra information.
• Examples
• Maximum payload host is willing to accept can be
  advertised during connection setup.
• Window scale factor that allows sender and
  receiver to negotiate larger window sizes.

34
UDP

• Provides connection-less, unreliable service.
• No delivery guarantees.
• No ordering guarantees.
• No duplicate detection.
• Low overhead.
• No connection establishment/teardown.
• Suitable for short-lived connections.
• Example client-server applications.

35
UDP Segment Format
0 15
31
Destination port
Source port
Length
Checksum
Data
Source and destination ports identify the end
points. Length 8-byte header data. Checksum
optional if not used, set to zero.
36
TCP and UDP

• TCP provides end-to-end communication. It takes
  care of reliable, error-free transfer of data,
  and in-sequence delivery
• UDP has less overhead compared to TCP, but does
  not guarantee transfers
• TCP is preferred to transfer files
• UDP is preferred to transfer audio/video streams
• In real-time streaming, we cannot afford the
  delay consequent to packet retransmission
• Both protocols support multiplexing, i.e. they
  allow several distinct streams of data between
  two hosts

37
The Domain Name System (DNS)
38
DNS

• IP addresses are not easy to remember.
• The Domain Name System (DNS) maps IP addresses to
  host names.
• Host name is formed by machine name followed by
  domain name.
• Host_name.domain_name

39
DNS (contd)

• The domain_name is formed by the institutional
  site name and the Top-Level Domain name (TLD).
• So the host name is of the formmachine_name.Ist_s
  ite_name.TLD_name
• Examples
• sundance.ucsc.edu
• soe.ucsc.edu (alias for sundance.ucsc.edu)
• italia.cse.ucsc.edu
• helios.jpl.nasa.gov
• www.cnn.com

40
TLD

• TLD names identify organization types or country
  codes.
• Examples
• .com Commercial org. .au Australia
• .edu Educational site in US .ca Canada
• .gov Government site in US .fr France
• .mil Military organization in US .de Germany
• .net Network site .uk Great Britain
• .org Nonprofit organization .it Italy
• .es Spain
• Countries define their own internal hierarchy
  (e.g., .ac.uk, .edu.au)

41
DNS (contd)

• Organizations can create any internal DNS
  hierarchy.
• Authority for creating new subdomains within a
  domain name is delegated to each domain.
• Administration of ucsc.edu has authority to
  create cse.ucsc.edu and need not contact any
  central naming authority.

42
Example of DNS Hierarchy
43
DNS Name Space

• DNS names are managed by a hierarchy of DNS
  servers.
• Hierarchy is related to DNS domain hierarchy
• Root server at top of tree knows about next level
  servers.
• Next level servers, in turn, know about lower
  level servers.

44
Example of DNS Hierarchy
45
Example of DSN Hierarchy (contd)
46
Choosing DNS Server Architecture

• Small organizations can use a single server.
• Easy to administer.
• Inexpensive.
• Large organizations often use multiple servers.
• Reliability through redundancy.
• Improved response time through load sharing.

47
Name Resolution

• Resolving a name means mapping the host name to
  the IP address.
• Reverse mapping is also possible.
• A client computer calls a DNS server for name
  resolution
• DNS request contains name to be resolved.
• DNS reply contains IP address for name in request.

48
Using DNS Servers

• Each DNS server is the authoritative server for
  the names it manages.
• If request contains name managed by receiving
  server, that server replies directly.
• Otherwise, request is forwarded to the
  appropriate authoritative server.
• DNS request is originally sent to root server,
  which points at next server to use
• Eventually, the authoritative server for the DNS
  name in the request is located and IP address is
  returned.