Chapter 5: Network and Transport Layers

1
Chapter 5 Network and Transport Layers
2
Outlines

• Network Protocols and TCP/IP
• Networking Addressing
• Routing
• Network flow control and QoS

3
Network Protocols and TCP/IP
4
Transmission Control Protocol/ Internet Protocol
(TCP/IP)

• The Transmission Control Protocol/ Internet
  Protocol (TCP/IP) was developed for the U.S. Dept
  of Defenses Advanced Research Project Agency
  Network (ARPANET) in 1974.
• TCP/IP allows reasonable efficient and error-free
  transmission.

5
TCP/IP

• TCP/IP has two parts
• TCP - performs packetizing TCP is only active at
  the sender and receiver.
• IP - performs routing and addressing.
• A typical TCP packet has 192-bit (24-byte) header
  of control information.

6
TCP/IP

• Two forms of IP are currently in use
• IPv4 also has a 192-bit (24-byte) header.
• IPv6 has a 320-bit (40-byte) header.
• The primary reason for the increase in packet
  size is an increase in the address size from 32
  bits to 128 bits, due to the dramatic growth in
  the usage of the Internet.
• The size of the message field depends on the data
  link layer protocol used. TCP/IP is commonly
  combined with Ethernet.

7
TCP Packet
1
2
3
4
5
6
7
8
9
10
11
User Data
1 Source ID 16 bits 2 Destination ID 16
bits 3 Sequence number 32 bits 4 ACK number 32
bits 5 Header length 4 bits 6 Unused 6
bits 7 Flags 6 bits 8 Flow control 16
bits 9 CRC 16 16 bits 10 Urgent pointer 16
bits 11 Options 16 bits
8
IP Packet version
IP4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1 Version number 4 bits 2 Header length 4
bits 3 Type of Service 8 bits 4 Total length 16
bits 5 Identifiers 16 bits 6 Flags 3
bits 7 Packet offset 13 bits 8 Hop limit 8 bits
9 Protocol 8 bits 10 CRC 16 16 bits 11 Source
address 32 bits 12 Destination Address 32
bits 13 Options varies 14 User
data varies 15 Flow name 24 bits 16 Next
header 8 bits
IP6
1
15
4
16
8
11 (128 bits)
12 (128 bits)
14
9
History of IPng Effort

• By the Winter of 1992 the Internet community had
  developed four separate proposals for IPng. These
  were "CNAT", "IP Encaps", "Nimrod", and "Simple
  CLNP". By December 1992 three more proposals
  followed "The P Internet Protocol" (PIP), "The
  Simple Internet Protocol" (SIP) and "TP/IX". In
  the Spring of 1992 the "Simple CLNP" evolved into
  "TCP and UDP with Bigger Addresses" (TUBA) and
  "IP Encaps" evolved into "IP Address
  Encapsulation" (IPAE).
• By the fall of 1993, IPAE merged with SIP while
  still maintaining the name SIP. This group later
  merged with PIP and the resulting working group
  called themselves "Simple Internet Protocol Plus"
  (SIPP). At about the same time the TP/IX Working
  Group changed its name to "Common Architecture
  for the Internet" (CATNIP).
• The IPng area directors made a recommendation for
  an IPng in July of 1994 RFC 1752.
• The formal name of IPng is IPv6

10
Why Need IPv6?

• Internet Growth
• Network numbers and size
• Traffic management
• Quality of Services (QoS)
• Internet Transition
• Routing
• Addressing
• No question that an IPv6 is needed, but when

11
Other Protocols

• Internetwork Packet Exchange/Sequenced Packet
  Exchange (IPX/SPX)
• Developed by Xerox in the 1970s. It is primary
  network protocol used by Novell NetWare. Novell
  has replaced IPX/SPX with TCP/IP.
• X.25
• ITU-Ts standard for WAN. Mature standard. Seldom
  used in north America.
• System Network Architecture (SNA)
• IBM developed SNA in 1974. It is used on IBMs
  mainframes. It is hard to integrate SNA with
  other networks.

12
The Message Field Size

• Maximum Ethernet packet size 1492
• TCP message field
• 1492 - 24 (TCP header) - 24 (IPv4 header) 1444

13
Addressing
14
Types of addresses
Address Example Software Example
Address Application Layer Web browser ike.ba.ttu
.edu (also called domain name) Network
Layer TCP/IP 129.118.49.189 Data Link
Layer Ethernet 00-A0-C9-96-1D-90
15
Addressing

• The network layer determines the best route
  through the network to the final destination.
• Based on this routing, the network layer
  identifies the data link layer address of the
  next computer to which the message should be
  sent.

16
Assigning Addresses

• In general, the data link layer address is
  permanently encoded in each network card, and as
  part of the hardware that cannot be changed.
• Network layer addresses are generally assigned by
  software. Every network layer software package
  usually has a configuration file that specifies
  the network layer address for that computer.

17
Assigning Addresses

• Application layer addresses (or server addresses)
  are also assigned by a software configuration
  file. Virtually all servers have an application
  layer address, but most client computers do not.
• Network layer addresses and application layer
  addresses go hand in hand. ike.ba.ttu.edu - means
  129.118.49.189 at the network layer.)

18
How IP Addresses Distributed

• Internet Corporation for Assigned Names and
  Numbers (ICANN) oversees the Internet Assigned
  Numbers Authority (IANA) and controls how the
  Net's 4.29 billion IP addresses are used.
• IANA distributes address space to three
  geographically diverse Regional Internet
  Registries (RIRs) and encourage three RIRs to
  operate so that addresses remain unique, are
  mapped efficiently, and are treated as a precious
  resource.
• Three RIRs dole out available pools of IP based
  on a shared criteria. All deploy numerical
  address space to ISPs, local registries, and in
  some cases small users.

19
IP Address Allocation
IANA
InterNIC America
RIPE Europe
APNIC Asia
National
Regional
Consumer
20
Three RIRs

• American Registry for Internet Numbers (ARIN)
• Reseaux IP Europeen (RIPE)
• Asia Pacific Network Information Centre (APNIC)

21
Internet Addresses

• InterNIC is responsible for network layer
  addresses (IP addresses) and application layer
  addresses or domain names (www.ttu.edu).
• There are five classes of Internet addresses.
• Classes A, B, and C are available to
  organizations
• Class D and E are reserved for special purposes
  and are not assigned to organizations.

22
Internet Address Classes

• Class A (/8 address)
• The first digit is fixed, ranging 1-126 (01-7E),
  16 million addresses
• 127.x.x.x is reserved for loopback
• Class B (/16 address)
• First two bytes are fixed with the first digit
  ranging 128-191 (80-BF), 65,000 addresses.
• Class C (/24 address)
• First 3 bytes are fixed, with the first digit
  ranging 192-223 (C0-DF), 254 addresses.
• Class D E
• The first digit is 224-239 (E0-EF) and 240-255
  (F0-FF) respectively.
• Reserved for special purposes and not available
  to organizations.

23
Internet Address Classes
Ranges of the first byte for different classes
224 239
126
128
191
192 223
1
240 255
1/2
1/4
1/8
1/16
1/16
Class B
Class A
Class D Class E
Class C
Class A 0xxxxxxx Class B 10xxxxxx.xxxxxxxx Class
C 110xxxxx.xxxxxxxx.xxxxxxxx Class D
1110xxxx.xxxxxxxx.xxxxxxxx Class E
1111xxxx.xxxxxxxx.xxxxxxxx
Note The IP addresses with the first byte as 0
and 127 are reserved
24
Internet Address Classes

• of Addresses
• Class Available Addr-Structure Example
  Available
• Class A 16 million First byte
  fixed 50.x.x.x 127
• Organization assigns
• last three bytes
• Class B 65k First two bytes
  fixed 128.192.x.x 16k
• Organization assigns
• last two bytes
• Class C 254 First three bytes
  fixed 192.1.56.x 2 millions
• Organization assigns
• last byte

25
Internet Addresses

• The Internet is quickly running out of addresses.
  Although there are more than 1 billion possible
  addresses, the fact that they are assigned in
  sets (or groups) significantly restricts the
  number of usable addresses.
• The IP address shortage was one of the reasons
  behind the IPv6, providing in theory, 3.2 x 1038
  possible addresses.
• How to apply for IP address?

26
Subnets

• Assign IP addresses to specific computers so that
  all computers on the same local area network have
  a similar address.
• Each LAN that is logically grouped together by IP
  number is called a TCP/IP subnet.
• Benefit
• allows it to be connected to the Internet with a
  single shared network address
• an necessary use of the limited number of network
  numbers
• Overload Internet routing tables on gateways
  outside the organization

27
Gateway
146.7.11.1
128.192.254.2
28
Subnet Mask

• Subnet mask enables a computer to determine which
  computers are on the same subnet. This is very
  important for message routing.
• E.g.
• IP address 129.118.49.189
• Subnet mask 255.255.255.0
• IP address 129.118.49.x is for the
  computers in the same subnet

29
Subnet

• Subnet with partial bytes addresses.
• E.g. 129.118.49.1 to 129.118.49.126
• Subnet mask 255.255.255.128
• Subnet address 129.118.49.0
• Subnet broadcast address 129.118.49.127

30
Subnet
IP address 129.118.49.111 1000 0001.0111
0110.0011 0001.0110 1111 Subnet
mask 255.255.192.0 1111 1111.1111 1111.1100
0000.0000 0000 The IP prefix 1000 0001.0111
0110.00 Destination IP 129.118.51.254 1000
0001.0111 0110.0011 0011.0110 1111 Destination
IP 128.83.127.1 1000 0000.0101 0011.0111
1111.0000 0001
31
Subnet Mask Template
Broadcast Address
150.1.0.0
255
255
0
0
Host Address
150
1
128 64 32 16 8 4 2 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
1 0 0 1 0 1 1 0
0 0 0 0 0 0 0 1
Network IDClass B
128
Mask Numbers
Possible Subnet Address
32
Dynamic Addressing

• An address assignment problem
• Each time the computer is moved, or its network
  is assigned a new address, the software on each
  individual computer must be updated.
• Solution dynamic addressing
• With this approach, a server is designated to
  supply a network layer address to a computer each
  time the computer connects to the network.

33
Dynamic Addressing

• Two standards for dynamic addressing are commonly
  used in TCP/IP networks
• Bootstrap Protocol (bootp) for dial-up networks
  (1985)
• Dynamic Host Control Protocol (DHCP) for
  non-dial-up networks (1993)

34
Dynamic Addressing

• The Bootp or DHCP server can be configured to
  assign the same network layer address to the
  computer each time it requests an address or it
  can lease the address to the computer by picking
  the next available network layer address from
  a list of authorized addresses.
• Dynamic addressing greatly simplifies network
  management in non-dial-up networks too.

35
Address Resolution

• Address resolution
• The sender translates the application layer
  address (or server name) of the destination into
  a network layer address and in turn translates
  that into a data link layer address.
• Two approaches used in TCP/IP
• Server address resolution
• Data link layer address resolution.

36
Server Name Resolution

• Domain Name Service (DNS)
• Used for translating application layer addresses
  into network layer addresses.
• InterNIC
• Keeps the name and IP addresses of the name
  server that will provide DNS information for your
  address classes.

37
Domain Name System

• 32-bit IP addresses have two drawbacks
• Routers cant keep track of every network path
• Users cant remember dotted decimals easily
• Domain names address these problems by providing
  a name for each network domain (hosts under the
  control of a given entity)

38
DNS Database

• Hierarchical database containing name, IP
  address, and related information for hosts
• Provides name-to-address directory services
• Key features
• Variable-depth hierarchy. Unlimited levels
• Distributed database. Scattered throughout the
  Internet and private intranet.
• Distribution controlled by the database.
  Thousands of separately managed zones managed by
  separate administrators

39
Server Name Resolution

• Server address resolution process
• TCP/IP sends a special TCP-level packet to the
  nearest DNS server asking for the requesting
  computer the IP address that matches the Internet
  address provided.
• If the DNS does not have the answer for the
  request, it will forward the request to another
  DNS.
• This is why it sometimes takes a long time to
  access certain sites.
• IP addresses are then temporarily stored in a
  server address table.

40
Data Link Layer Address Resolution

• In order to actually send a message, the network
  layer software must know the data link layer of
  the destination computer.
• In the case of a distant computer, the network
  layer would route the message by selecting a path
  through the network that would ultimately lead to
  the destination.

41
Data Link Layer Address Resolution

• The process
• TCP/IP software sends a broadcast message (using
  Address-Resolution-Protocol or ARP) to all
  computers in its subnet requesting the data link
  layer address.
• The computer with the right IP address responds
  with its data link layer address
• The message is sent to the destination computer

42
Routing
43
(No Transcript)
44
Routing

• There are many possible routes or paths a message
  can take to get from one computer to another.
• Routing
• The process of determining the route or path
  through the network that a message will travel
  from the sender to the receiver.
• Routing table
• The routing information on each router, which
  specifies how message will travel through the
  network.
• Types of routing
• Centralized routing
• Decentralized routing Static routing, Dynamic
  routing

45
Routing
46
Routing Table for Computer B

• Destination Route
• A A
• C C
• D A
• E E
• F E
• G C

47
Static Routing

• Static Routing
• The routing table is developed by the network
  manager, and changes are made only when computers
  are added or removed from network.

48
Dynamic Routing

• Dynamic Routing (adaptive routing)
• An initial routing table is developed by the
  network manager, but is continuously updated by
  the computers themselves to reflect changing
  network conditions, such as network traffic.
• Used when there are multiple routes through a
  network and it is important to select the best
  (or fastest) route, in order to route messages
  away from traffic on busy circuits.

49
Dynamic Routing

• Commonly used dynamic routing protocols
• Routing Information Protocol (RIP) - used by the
  network manager to develop the routing table.
• Border Gateway Protocol (BGP). A dynamic exterior
  routing protocol for the Internet.
• Internet Control Message Protocol (ICMP) - used
  on the internet with TCP/IP.
• Open Shortest Path First (OSPF) uses the number
  of computers in a route as well as network
  traffic and error rates to select the best route.
• Enhanced Interior Gateway Routing Protocol
  (EIGRP) a dynamic link state interior routing
  protocol and commonly used inside an organization.

50
Dynamic Routing

• Routing Information Protocol (RIP)
• When new computers are added, it counts the
  number of computers in the possible routes to the
  destination and selects the rout with the least
  number.
• Computers using RIP send broadcast messages every
  minute or so to announce routing state.
• It is used by TCP/IP and IPX/SPX.

51
Dynamic routing

• Border Gateway Protocol (BGP)
• A dynamic routing protocol used on the Internet
  to exchange routing information between
  autonomous systems the large sections of the
  Internet. It is seldom used inside companies
• Large, complex and hard to administer

52
Dynamic Routing

• Internet Control Message Protocol (ICMP)
• Uses both broadcast messages and the messages to
  specific computers to exchange routing
  information
• Only used by TPC/IP

53
Dynamic Routing

• Open Shortest Path First (OSPF)
• More efficient than RIP because it normally
  doesnt use broadcast messages. Instead it
  selectively sends status update messages directly
  to selected computers
• Used by TCP/IP

54
Dynamic routing

• Enhanced Interior Gateway Routing Protocol
  (EIGRP)
• A dynamic link state interior routing protocol
  developed by CISCO
• Commonly used inside an organization
• Computers/routers store their own routing table
  and their neighbors routing tables

55
Dynamic Routing

• Two drawbacks to Dynamic Routing.
• It requires more processing by each computer in
  the network than centralized or static routing.
• The transmission of status information wastes
  network capacity.

56
Connectionless vs. Connection-Oriented Routing

• Two ways a group of packets can be routed
• Connectionless routing
• Each packet is treated separately and makes its
  own way through the network.
• Connection-Oriented routing
• Sets up a virtual circuit between the sender and
  receiver. Appears to use point-to-point
  circuit-switching, but actually uses
  store-and-forward.
• Has greater overhead than connectionless, due to
  the routing information.

57
Connectionless vs. Connection-Oriented Routing

• Virtual Circuit
• Appears to the application software to use a
  point-to-point circuit
• The network layer makes one routing decision and
  all packets follow the same route

58
Connectionless vs. Connection-Oriented Routing

• TCP/IP vs. UPD/IP
• TCP/IP is used for connection-oriented routing
• TCP establishes the virtual circuit and IP routes
  the messages.
• UDP/IP is used for connectionless routing
• The TCP packet is replaced with a User Datagram
  Protocol (UDP) packet.

59
Multicast

• Unicasting
• The usual transmission between two computers.
• Broadcasting
• Sending messages to all computers on a LAN or
  subnet.
• Multicasting
• Sending the same message to a group of computers
  temporarily in a class D IP address.

60
Broadcast
Individual transfers
Clients
Host
61
Multicast
Could be one packet that all receive
or replicated by routers in the network
Data replicated by the network
Clients
Host
Multicast Infrastructure
One transfer
62
Multicast

• Computers wishing to participate in a multicast
  send a message to the sending computer or some
  other computer performing routing along the way
  using a special type of TCP-level packet called
  Internet Group Management Protocol (IGMP).
• Each multicast group is temporarily assigned a
  special Class D IP address to identify the group,
  thus allowing a restricted broadcast of messages
  to this specific group.

63
TCP/IP
TELNET FTP SMTP DNS SNMP
DHCP
Application Presentation Session
RIP
RTP RTCP
Transmission Control Protocol
User Datagram Protocol
Transport
OSPF
ICMP
IGMP
Internet Protocol
Network
ARP
Data link Physical
Ethernet
Token Bus
Token Ring
FDDI
64
Flow control and QoS
65
Quality of Service

• Quality of Service (QoS)
• The idea that transmission quality (rates, error
  rates, bandwidth and jitter) can be measured,
  improved, and, to some extent, guaranteed in
  advance.
• QoS routing
• A special type of connection-oriented dynamic
  routing in which different messages or packets
  are assigned different priorities.

66
Categories of Traffic

• Elastic traffic, such as FTP, email, etc
• Allow fluctuating bandwidth, the total
  transmission time is important
• The data must correctly transmitted
• Service quality concerns mainly in transmission
  delay and error control.
• Real-time traffic, such as videoconferencing.
• Demands certain bandwidth with isochronous
  features
• Tolerates some level of errors.
• Service quality criteria include Throughput,
  Delay, Delay variation (jitter), and Packet loss.

67
Routing at Routers

• Bandwidth schedule
• First in first out
• Round robin
• Prioritization
• Queue management
• Packet discard policy
• Congestion control

Packet arrival
Packet forward
Packet Drop
68
Network Congestion

• What is traffic congestion?
• The buffer in a forwarding device overflows. This
  results packet losses and incur retransmission.
  The transmission will worsen the situation.
• Network congestion control is very important in
  flow management

69
Internet Flow Control

• Internet flow control algorithm
• Slow start, congestion avoidance
• Router queue management
• Random early detection (RED) for packet dropping
• Data flow scheduling
• FIFO, round robin, priority queueing, weighted
  fair queueing

70
Internet Flow Control

• Slow Start algorithm (RFC2001). To avoid router
  running out of space
• Two windows advertised window by receiver and
  congestion window by sender. The congestion
  window is flow control imposed by the sender,
  while the advertised window is flow control
  imposed by the receiver.
• The congestion window is initialized to one
  segment. Each time an ACK is received, the
  congestion window is increased by one segment.
  The sender can transmit up to the minimum of the
  congestion window and the advertised window.
• The sender starts by transmitting one segment and
  waiting for its ACK. When that ACK is received,
  the congestion window is incremented from one to
  two, and two segments can be sent.
• When each of those two segments is acknowledged,
  the congestion window is increased to four. This
  provides an exponential growth.
• At some point the capacity of the internet can be
  reached, and an intermediate router will start
  discarding packets. This tells the sender that
  its congestion window has gotten too large.

71
Internet Flow Control

• Congestion Avoidance (RFC2001)
• Sets congestion window to one segment.
• When congestion occurs (indicated by a timeout or
  the reception of duplicate ACKs), one-half of the
  current window size (the minimum of congestion
  window and the receiver's advertised window, but
  at least two segments) is saved as X.
• When new data is acknowledged by the other end,
  increase congestion window, but the way it
  increases depends on whether TCP is performing
  slow start or congestion avoidance. If congestion
  window is less than or equal to X, TCP is in slow
  start otherwise TCP is performing congestion
  avoidance.
• Slow start continues until TCP is halfway to
  where it was when congestion occurred (since it
  recorded half of the window size that caused the
  problem in step 2), and then congestion avoidance
  takes over.
• Congestion avoidance dictates that congestion
  window be incremented a linear growth of
  congestion window, compared to slow start's
  exponential growth.

72
Internet transmission services

• Best-effort services
• The Internet treats all packet equally.
• Integrated services (IntServ)
• IntServ refers to mechanisms that enable users to
  request a particular QoS for a flow of data.
• Differentiated Services (DiffServ)
• DiffServ Use type-of-service in IPv4 header to
  indicate the required service quality.

73
Integrated Services

• Routers require additional functionality to
  handle QoS-based service
• IETF is developing suite of standards to support
  this
• Two standards have received widespread support
• Integrated Services Architecture (ISA) To enable
  the provision of QoS support over IP-based
  Internet.
• Resource ReSerVation Protocol (RSVP)

74
Integrated Services Architecture

• Enables provision of QoS over IP-networks
• Features include
• Admission Control A new flow needs a reservation
  for QoS
• Routing Algorithm more parameters are considered
  other than just delay
• Queuing Discipline Queuing policy takes into
  account of different requirements
• Discard Policy Particularly for congestion
  management

75
Resource Reservation Protocol (RSVP)

• A tool for prevention of congestion through
  reservation of network resources
• Can be used in unicast or multicast transmissions
• Receivers (not senders) initiate resource
  reservations
• Operation
• Complexity is in multicast transmission
• RSVP uses two basic messages Resv and Path. In
  multicast, Resv messages generated by one of the
  multicast group receivers propagate upstream
  through distribution tree and create soft state
  in routers. Once it reaches the sender, hosts are
  enabled to set parameters for the first hop. Path
  is used to provide upstream routing information
  and sent from senders via the down stream tree to
  all receivers

76
Differentiated Services (DiffServ)

• Provides QoS based on user group needs rather
  than traffic flows
• Can use current IPv4 octets
• Service-Level Agreements (SLA) govern DiffServ,
  eliminating need for application-based assignment

77
IPv4 Type of Service Field

• Allows user to provide guidance on individual
  datagrams
• 3-bit precedence subfield
• Indicates degree of urgency or priority
• Queue Service Congestion Control
• 4-bit TOS subfield
• Provides guidance on selecting next hop
• Route selection, Network Service, Queuing
  Discipline

1
2
3
4
5
6
7
0
Precedence
TOS
0
78
DiffServ Domains
Border component
Host
Host
Interior component
79
DiffServ Operation

• Routers are either boundary nodes or interior
  nodes
• Interior nodes use per-hop behavior (PHB) rules
• Boundary nodes have PHB traffic conditioning

80
Token Bucket Scheme
Max Burstiness RT B
R Token replenishment rate B Bucket size
81
TCP/IP Configuration Information

• At least four pieces of information needed for a
  client computer TCP/IP configuration
• IP address
• Subnet mask
• Gateway IP address
• Domain name Server IP address

82
A TCP/IP Example
83
A TCP/IP Example

• How a client access a web server in the same
  subnet with a known address?
• How a client access a web server in a different
  subnet with a known address?
• How a client access a web server in the same
  subnet with an unknown address?

84
Sender
Receiver
Application Layer
Application Layer
HTTP
Request
HTTP
Request
Transport Layer
Transport Layer
HTTP
TCP
Request
HTTP
TCP
Request
Network Layer
Network Layer
HTTP
TCP
IP
Request
HTTP
TCP
IP
Request
Data Link Layer
Data Link Layer
HTTP
TCP
IP
Ethernet
Request
HTTP
TCP
IP
Ethernet
Request
Physical Layer
Physical Layer
85
Data transmission using TCP/IP and Ethernet
Ethernet packet header
IP packet
TCP packet
HTTP packet
User Data
Ethernet packet trailer
IP address
Data link layer address