Transcript Slide 1

Chapter 15: Networks
Data communication networks are set up in
two basic configurations.
The mesh configuration uses direct, point-topoint connections between each pair of
communicating devices.
While this approach guarantees a direct connection between any
pair of devices, the amount of cabling and I/O hardware required
at each device makes it impractical in most situations.
The alternative is to establish a broadcast network in which
network lines are shared.
Simple configurations
associated with the broadcast
approach include the ring and
bus topologies, at left and
right, respectively.
Chapter 15
Networks
Page 162
Local Area Networks
LANs are privately owned networks containing perhaps dozens of
devices.
Example: Ethernet
Computer
Computer
Bus Cable
Computer
Computer
Computer
Computer
Ethernet uses a bus configuration and a protocol for accessing the bus called
CSMA/CD.
• CS - Carrier Sense: Each machine constantly listens to the traffic that’s
passing by on the bus.
• MA - Multiple Access: Every machine has equal access to the communication
medium (i.e., the bus).
• CD - Collision Detection: Each machine is capable of detecting
whether its transmitted message “collided” with that of another
Chapter 15
Networks
machine, thus corrupting the message and forcing a retransmission.
Page 163
Ethernet Carrier Sense: Incoming Traffic
As messages pass by, each machine examines the first few bits of
the message, the “address” of the message, and determines
whether or not the message is intended for itself.
Bus Cable
message
Computer
Computer
Computer
Computer
Computer
Computer
message
When the destination machine sees its
address in the message’s prefix, it copies
the passing message as the original
proceeds down the bus.
Chapter 15
Networks
Page 164
Ethernet Carrier Sense: Outgoing Traffic
When a machine wants to transmit a message, it waits a systemdependent amount of time, using its carrier sense to see if any
traffic appears on the bus. If not, it transmits its message and
“hopes” that no collision occurs.
Bus Cable
Computer
Computer
Computer
Computer
Computer
message
Computer
When the transmitting machine detects
no passing traffic for a certain time
interval, it places its message on the bus.
Chapter 15
Networks
Page 165
Ethernet Carrier Sense: Colliding Traffic
After a machine transmits a message, it continues to use its carrier
sense to see if there’s any difference between its transmitted
message and the message on the bus. If so, it interprets the
problem as a collision.
Bus Cable
Computer
Computer
Computer
Computer
When each transmitting machine
detects that its message has been
corrupted, it waits a random interval of
time, and then retransmits.
Computer
Computer
Chapter 15
Networks
Page 166
A LAN Alternative: Token Ring
Collisions tend to exacerbate traffic problems on congested
Ethernet LANs, so an alternative to the CSMA/CD approach is
desirable.
Ring
Computer
Computer
Computer
message
Computer
TOKEN
Computer
Computer
Computer
message
TOKEN
message
Computer
Computer
Computer
Computer
Computer
Computer
Token Ring uses a ring topology, with a specially formatted “token” message
perpetually traversing the ring.
When a machine wishes to transmit, it merely waits for the token to arrive,
removes it, and transmits its message.
When its message returns to the transmitting machine, it’s removed
from the ring, and the machine places the token back on the ring.
Chapter 15
Networks
Page 167
Wide Area Networks
WANs are large collections of smaller networks, with special
interconnection devices known as “routers” to make adjacent
sub-networks compatible.
Router
Router
Router
Router
Router
Router
Chapter 15
Networks
Page 168
Switching
Data communications is characterized by two switching
technologies.
Technology #1: Circuit Switching
Once the circuit is established, it is
maintained until one of the endstations
terminates the connection.
Chapter 15
Networks
Page 169
Problems with Circuit Switching
Circuit switching has one great advantage:
once established, the circuit is dedicated,
i.e., a communication line is completely
open until formally terminated.
However, this approach has a number
of serious problems:
• Many networked applications don’t require a dedicated circuit, so reserving a
communication line until an endstation formally terminates it can represent a
serious waste of resources.
• The route originally selected for the circuit may be optimal to begin with, but
may prove to be suboptimal as the communication continues.
• An entire, end-to-end path must be found and reserved before any
communication is allowed between the two endstations; this is definitely not
conducive to many modern applications (e.g., Web surfing, videoconferencing).
• Transmission errors are propagated all the way to the destination,
requiring retransmission across the entire network.
Chapter 15
Networks
Page 170
An Alternative to Circuit Switching
Technology #2: Packet Switching
The source’s message is broken into manageable “packets” that are transmitted
to the destination individually, not necessarily along the same path.
Chapter 15
Networks
Page 171
The Pros and Cons of Packet Switching
Packet switching remedies circuit
switching’s principal problems:
• Lines aren’t dedicated, so their
utilization is higher.
• Messages are “packetized”, so linesharing is reasonably fair.
• Routing may be dynamic, i.e., an
alternate route may be chosen when
traffic patterns change.
• The entire route does not have to be
chosen prior to sending any data.
• Errors aren’t propagated end-to-end.
However, packet switching does have its own set of problems:
• Switches must be programmed to make sophisticated routing decisions.
• Switches must manage memory for queued packets that await forwarding.
• Packets must be prefixed with control headers, increasing overhead.
Chapter 15
Networks
• Endstations must deal with missing packets and out-of-order packets.
Page 172
• Without a dedicated circuit, transmission times become unpredictable.
Multiplexing
To more efficiently utilize a physical medium, multiple higherlevel connections might “share” the medium simultaneously.
Frequency-Division Multiplexing
The spectrum of frequencies transmittable via the physical medium is divided into
several channels (e.g., cable TV).
Time-Division Multiplexing
Each transmitter is allocated a periodic time interval in which to transmit.
Chapter 15
Networks
Page 173
Code Division Multiplexing
Frequency Division
Multiplexing
Everyone gets to
talk at the same
time, but only
across their narrow
channels.
(Commonly used
with copper
cables)
Time Division
Multiplexing
Everyone gets to talk
using the entire
bandwidth, but they
have to take turns
talking.
(Commonly user with
fiber optics)
Code Division Multiplexing
Everyone gets to talk simultaneously, using the entire bandwidth!
They do this by coding their transmissions in a unique fashion (as if
every pair were speaking a different language, and each other
language merely sounds like background noise).
(Commonly used with wireless communications)
Chapter 15
Networks
Page 174
Modems
Digital data must be modulated into analog signals if it’s going
to be transmitted across an analog medium.
After transmission, it must be demodulated back into its original
digital form.
Cables between the workstation
and the modem and between the
modem and the telephone jack
Dial-in pool of 96 modems:
Six terminal servers, each
connecting 16 modems to a LAN
Chapter 15
Networks
Page 175
Modulation and Demodulation
Modulation of the digital signals
m1 and m2 is effected at left via
products with trig functions.
cos(ct)
m1(t)
X
cos(ct) m1(t) +
sin(ct) m2(t)

m2(t)
X
sin(ct)
Upon arrival, the message is
demodulated below via trig
multiplications, trig identities,
and high-frequency filtering.
2cos(ct)
m1(t)+cos(2ct) m1(t)
+sin(2ct) m2(t)
X
cos(ct) m1(t) +
sin(ct) m2(t)
X
m2(t)+cos(2ct) m2(t)
+sin(2ct) m1(t)
FILTER OUT
THE HIGH
FREQUENCIES
m1(t)
m2(t)
2sin(ct)
Chapter 15
Networks
Page 176
Modem Interfaces
The EIA-232D connector connects
a modem to a computer.
• It uses a 25-pin connector, with
each pin having a different
meaning.
• The meaning of each pin is
activated by applying a voltage
across it.
For example...
Pin 20: Data Terminal Ready
(i.e., “Hey! The computer’s ready to
send something to the modem!”)
Chapter 15
Networks
Page 177
Network Protocol Layers
In an effort to simplify networks, they are often organized as
layered hierarchies of protocols, with hardware-intensive
protocols on the bottom and user applications at the top.
Layer 5
Layer 5
Layer 4
Layer 4
Layer 4
Layer 4
Layer 3
Layer 3
Layer 3
Layer 3
Layer 2
Layer 2
Layer 2
Layer 2
Layer 2
Layer 1
Layer 1
Layer 1
Layer 1
Layer 1
Source
Host
Intermediate
Router
Low-Level
Bridge
Intermediate
Router
Destination
Host
Physical Medium
Chapter 15
Networks
Page 178
Communicating Via Layered Protocols
Consecutive network nodes only communicate directly at the
lowest (hardware) layer; to communicate at higher layers,
networking software inserts certain relevant data as headers
and trailers to the message coming from the source.
Layer 5
message
Layer 4
Layer 4
hdr4 message
hdr4 message
Layer 3
Layer 3
hdr3 hdr4 msgB
hdr3 hdr4 msgA
hdr3 hdr4 msgA
hdr3 hdr4 msgB
Layer 2
Layer 2
hdr2 hdr3 hdr4 msgA trl2
hdr2 hdr3 hdr4 msgB trl2
hdr2 hdr3 hdr4 msgA trl2
hdr2 hdr3 hdr4 msgB trl2
Layer 1
Layer 1
Chapter 15
Networks
Page 179
hdr1 hdr2 hdr3 hdr4 msgB trl2
hdr1 hdr2 hdr3 hdr4 msgA trl2
Layered Protocol Models
Several models have been developed to implement protocol
hierarchies for networks.
Reference Model #1: Open Systems Interconnection (OSI)
Application
• End-user protocols, e.g., e-mail, file transfer, Web browsing,
network management, videoconferencing.
Presentation
• Format data according to negotiations between source and
destination; encrypt and decrypt messages.
Session
Transport
• Establish, maintain, and discontinue dialogues between a
source and a destination; synchronize data transfer.
• Detect and handle end-to-end transmission errors; alter
transmission rate when too much congestion is encountered.
Network
• Route messages from their source to their destination; reroute
traffic when heavy congestion is encountered.
Data Link
• Detect and handle transmission errors between consecutive
network devices; control access to the shared medium.
Physical
• Transmit bits across physical medium; determine frequencies
to use when transmitting; specify role of connector pins.
Chapter 15
Networks
Page 180
Reference Model #2: Transmission Control Protocol/Internet Protocol (TCP/IP)
Application
• Analogous to OSI’s Application Layer
TCP
• Analogous to OSI’s Transport Layer
IP
• Analogous to OSI’s Network Layer
Lower Levels • Analogous to OSI’s Physical and Data Link Layers
Chapter 15
Networks
Page 181
The IP Protocol
The Internet Protocol was designed for three primary purposes:
1. Define the data format to be used by messages
travelling through TCP/IP networks.
2. Route the data through the Internet by
selecting appropriate paths.
3. Process packets, generate error messages,
and discard packets in such a way to ensure
“unreliable” packet delivery.
Chapter 15
Networks
Page 182
The IP Header
Version HdrLen
Service Type
Identification
Time To Live
Total Length
Flags
Protocol
Fragment Offset
Header Checksum
Source IP Address
Destination IP Address
Options & Padding (if any)
Version: The version of IP used to create the packet, used by nodes to process it correctly.
HdrLen: The length of the header in 32-bit words (because the Options field has no fixed size).
Service Type: Six bits to represent the relative priority and delay sensitivity of the packet.
Total Length: The length of the entire packet in bytes (16-bit field means a 65,535-byte max).
Identification: All fragments of the same packet have the same ID number.
Flags: Don’t-Fragment flag and More-Fragments flag.
Fragment Offset: Offset from start of packet (in bytes) of current fragment.
Time To Live: Length of time (in seconds) the packet may stay in the Internet.
Protocol: Global ID # of the protocol used to create the packet (e.g., TCP).
Header Checksum: Error-checking sum of all of the 16-bit values in the header.
Source IP Address: 32-bit IP address of the packet’s original source.
Destination IP Address: 32-bit IP address of the packet’s final destination.
Options & Padding: Options include: No-operation-just-align; Military-security-application;
Loose-source-routing; Record-route; Strict-source-routing; Record-internet-timestamps.
Chapter 15
Networks
Page 183
IP Addresses
There are three principal classes of IP addresses, with all
endstations on the same network given a common prefix:
CLASS A - for networks with at least 216 (65536+) endstations:
Host ID
0 Network ID
CLASS B -for networks with between 28 and 216 (256-65535) endstations:
10
Network ID
Host ID
CLASS C - for networks with less than 28 (0-255) endstations:
110
Network ID
Host ID
Chapter 15
Networks
Page 184
The Domain Name System (DNS)
A hierarchical system of domains and
subdomains has been established to
permit stations to communicate with
other stations by “name”.
A station contacts its server, who knows
the location of the required domain
server, who knows the location of the
required subdomain server, etc., until
the required endstation is located,
whereupon its IP address is returned.
Chapter 15
Networks
Page 185
Electronic Mail
User Sends Mail
User
Interface
User Reads Mail
Outgoing
Mail Spool
Area
Mailboxes
for Incoming
Mail
Client
TCP Connection
(Background
Transfer) for Outgoing Mail
Server
(To Accept
Mail)
TCP Connection
for Incoming Mail
The background transfer process sweeps through the spool area
periodically (typically, twice an hour). Whenever it finds an
undelivered message or whenever a user deposits new outgoing
mail, the background process attempts delivery.
Alias
Database
User Sends Mail
User
Interface
User Reads Mail
Alias Expansion
& Forwarding
Mailboxes for
Incoming Mail
Outgoing
Mail Spool
Area
Client
TCP Connection
(Background
Transfer) for Outgoing Mail
Server
(To Accept
Mail)
TCP Connection
for Incoming Mail
Rather than resorting to the Domain Name System for each
outgoing message, and to accommodate locally used aliases for
incoming messages, most systems provide mail forwarding
software with a mail alias expansion mechanism.
Chapter 15
Networks
Page 186
File Transfer Protocol (FTP)
Client System
control
process
Server System
One active TCP connection
before and after data
transfer, just for control.
Operating System
control
process
Operating System
TCP/IP Internet
Client System
data
transfer
control
process
Server System
Two active TCP
connections during data
transfer, one for control
and one for data.
Operating System
control
process
data
transfer
Operating System
TCP/IP Internet
Chapter 15
Networks
Page 187
Firewalls
To ensure the security of a private network,
“firewall” programs have been developed.
A common approach is to filter incoming and
outgoing packets based upon header
information, and to use an application gateway
to inhibit application-specific traffic.
packet
Incoming packets for bad
address/port combinations are
rejected (e.g., no outsider can
“finger” an internal site).
packet
Incoming or outgoing
packets are rejected on
the basis of size or
payload info.
packet
My Secure
Network
packet
Outgoing packets for bad
address/port combinations are
rejected (e.g., no insider can “http”
an external site).
Chapter 15
Networks
Page 188