Transcript Network - Directory

Chapter 3 Networks and Telecommunications
I. Why Networking
1. Purpose of networking: Connect computers for
the purpose of sharing information and resources.
• Data sharing (p.81)
– Master copies of data files reside on a computer
elsewhere on the network and users access the
master copy to do their work.
– It allows multiple users access the same file
simultaneously and it is able to merge multiple
updates to keep a single master copy consistent
and correct.
• Hardware sharing: (p.85) Printers, scanners,
storage spaces, special processors, and other devices
can be attached to a network. For many businesses,
this capability alone justifies the costs involved in
networking.
• Software sharing: (p.86) A network version of
software is stored in one computer (file server), and
users load the software package into the RAM of
their computers when they want to use it.
– It saves storage spaces and time for upgrading
since only one copy has to be installed.
– The number of users may exceed the number of
licenses if they do not use the software at the
same time.
2. Business reasons for data sharing
• Managers can see data immediately as it is
collected or updated.
• POS (point of sales)
• Decision support:
– speed of decision process
– error prevention
– updated information
• Information to circulate among users: E-mail,
Bulletin Board, Newsgroup and Chat Room.
• Calendar and appointment scheduling
• Teamwork and Groupware (ICQ)
• Data backup by system administrator
II. Components of a network
1. LAN and WAN
• Local Area Network (LAN): a small network
that encompasses a limited distance (normally
no more than 1,000 feet and one or two
buildings).
• Wide Area Network (WAN): A large network
that connects multiple groups of users in
multiple locations, e.g., enterprise networks. It
spans distance measured in miles and involves
links that are controlled by public carriers.
2. Four basic components of a network (p.88-p.96)
• Computers: two basic roles in networks
– Server: a computer that responds to request
by providing the requested information and
shares its resource across the network.
– Client: A computer that requests information
and accesses shared resources.
– Client/server network: Certain computers
function more or less exclusively as servers
while users’ computers function more or less
exclusively as clients. Servers have more CPU
power and storage capacity. The message
traffic between servers and clients is heavy.
– Peer-to-peer network: Computers function as
either servers or clients and at more or less the
same level of capacity. Older machines may
slow down too much while working as servers.
• Transmission Media
(cabling and wireless networking)
– Transmission capacity is typically measured in
Mbps (megabits per second).
Example:
Voice channel 300 Hz – 3400 Hz – its bandwidth is
3100 Hz. Suppose 1 bit per hertz of bandwidth, the
bit rate is 3100 bps.
• Transmission Media
– Types of cable
a) Coaxial cable (10 Mbps)
b) Twisted-pair cable (10 Mbps)
c) Fiber-optic cable (100 Mbps - 1 Gbps)
– Wireless networking (Radio, Micro, and Infrared
Waves)
a) Temporary connections into existing wired networks.
b) Contingency connections for existing wired networks.
c) Extend span beyond wired networks.
d) Travel with computers within certain limits.
Exercise: Sample Calculations
• Suppose you wanted to download the movie Titanic. In
compressed form it would contain about 4*109 bytes or
32*109 bits. If you had a typical 28.8 Kbps modem,
how long would it take to download the movie?
• Answer: 32*109 bits/ 28.8*103 bits/sec
= 1.11*106 sec or
= 1.11*106 sec/3.6* 103 sec/hour
= 308 hours or
= 308 hours/ 24 hours/day
=12.8 days
9
• Connection devices
– NIC (Network Interface Card): A physical interface between
your computer and the data network. It plugs into an adapter
slot inside the computer case. (Modem if telephone voice
channel is used.)
– Driver software: A device driver for NIC must be installed on
your computer.
– Hub: connect PC’s to network.
– Bridge & Switch: connect two network segments, filter and
forward packets.
– Router: connect multiple network segments or connect multiple
networks to form a complicated network such as Internet.
- Recognize node address and network address
- Repack data and select the best path
- Convert signals to interface different transmission
media and networks.
• Software
– Server network software: NOS (Network
Operating System) installed on servers. It is
a multi-user operating system.
– Client network software: the portion of NOS
installed on client computers for sending and
receiving information on network. It also can
convert data format for incoming and
outgoing messages.
– Application software: network version.
III. Networking standards (p.98)
• Ensure that computers from different vendors
can effectively work together.
• Meet up-to-date standards.
• Many standards and variations on standards.
History of Network Standards
• 1965 IBM research for LAN: only 20% information
flow goes outside.
• Protocols for LAN: Netware, Token Ring.
• 1969 ARPANET (now Internet) was a host-to-host
network between four universities. Protocols were Email, Telenet, and later FTP (File Transfer Protocol).
• 1980 UNIX NOS
• 1980 TCP/IP (p.99) (Transport Control Protocol / Internet
protocol) became the standard of Internet protocols.
• 1991 new language HTML (Hypertext Markup
Language) for pages and protocol HTTP (Hypertext
Transfer Protocol) for the transfer of data formats
between the server and the client.
• Protocol: “a set of rules that govern functional units
to achieve communication” – IEEE dictionary.
What are basic functions of a protocol?
• Segmentation and reassembly: Break up the data
message into packets and put packets back into their
original order.
• Encapsulation: add header and control information to
packets.
• Connection control: connection establishment, data
transfer, and connection termination.
• Ordered delivery: assign sequence numbers to packets
for reassembly on the receiving side.
• Flow control: manage the data flow such that buffer
memories do not overflow, but maintain full capacity.
• Error control: recovery of lost or errored packets.
* Adapted from “Practical Data Communications,” R.L. Freeman
What is OSI Networking Model?
• The Open Systems Interconnection (OSI) reference model was
proposed by the International Standards Organization (ISO).
• Networking can be broken into a series of tasks (layers):
7: Application (provide interface between application software and network)
6: Presentation (standardized data format, encryption/decryption, compression)
5: Session (setup, hold, end a communication – session)
4: Transport (data segmentation, flow control, error control, resequencing
packets)
3: Network (handle addressing messages; decide how to route transmissions –
quality, cost, and priority; router is working in this layer)
2: Data Link (package and send data frames – basic unit for network traffic “on
the wire”. Bridge and switch are working in this layer.)
1: Physical (covert bits into signals for outgoing messages and signals to bits for
incoming messages; hub is working in this layer)
• Each layer can be handled separately and its issues solved
independently.
15
IV. The Internet
1. Internet
• Internet: an international network of networks.
Computer – ISP – NSP – ISP - Computer (p.102)
• Email address
[email protected]
INDIVIDUAL OR
ORGANIZATION NAME
HOST
COMPUTER
Domain
Function
Country
16
• URL (Uniform Resource Locator)
– A URL is a location that indicates where a
resource on the Web can be found. In some
sense it is like the call number of a book.
– A URL contains three pieces of information
• the protocol used to access the resource
• the names of the host and the computer
where the resource is located (domain) and
the organization type
• the path name of the resource itself
Example: (next slide)
17
Parts of a
Uniform Resource Locator (URL)
hypertext
transfer
protocol
organization type
hypertext markup
language
http://www.winthrop.edu/ oncampus/academics/default.htm
Host name and
computer name
path
(directories and file name
on the web server)
• Intranets
– networks within an organization that use
the infrastructure and standards of the
Internet and WWW technology but have
“firewalls” (hardware and software) to
prevent outsiders from invading private
networks.
– Employees can go out but unauthorized
users cannot come in.
19
2. Three Approaches to Data Switching
• Circuit Switching: end-to-end connection during the
duration of a message.
• Message Switching: store and forward message.
• Packet Switching:
– Break down the message into short packages – packets
– Add header and tail so packets can be put back into the original
message on the receiving side.
– Packets can be stored, forwarded, and sent on diverse routes.
• What are advantages of packet switching?
– Efficient use of transmission links – diverse routes, store and
forward
– Near real time connectivity
– Highly reliable - if error occurs, only re-send the packet with
error rather than whole message.
– Highly survivable - if part of network is down, packets can be
stored and forwarded later.
3. Virtual Private Network
• It is a public network that provides services
to many companies.
• Your privacy is not protected by dedicated
line.
• Your privacy is protected by encryption
services provided by VPN provider.
21
V. Network Security
(Supplement)
1. What security services should network systems
provide?
• Confidentiality
• Access Control
• Integrity
• Non-repudiation
• Authentication
• Availability
Confidentiality and Eavesdropping
Eavesdropping: packet sniffing on net, in which
attackers read transmitted information, including
logon information and database contents.
History of Encryption and Decryption
• 1586 VIGENERE - paper and pencil, polyalphabetic
substitution cipher.
• 1920s - 1970s ENIGMA: substitution rings (rotors).
• 1975 US National Bureau of Standard (NBS): Data
Encryption Standard (DES) – a 56-bit key is no
longer considered to be very secure.
• 1990 Xuejia Lai and James Massey: IDEA – with a
128-bit key, approximately twice as fast as DES and
considerably more secure.
• 1977 Rivest, Shamir and Adleman: RSA public key
algorithm – a 2048-bit key is considered to be very
secure in the foreseeable future but is about 1000
times slower than DES.
• And more.
What are two major cryptographic methods?
• Conventional encryption: Message sender and
recipient share single secret key for encryption and
decryption. There are three basic operations:
- Substitution: replace bits with other bits.
- Transposition (permutation): arrange bits in a different
order
- XOR: 10110010  01110110 = 11000100
• Public-key encryption: Key owner generates a pair of
keys. One key, called public key (e), is made available
for anyone to get. Another key, called private key (d), is
kept by the owner. Message encrypted with one key
can be decrypted with another. The RSA algorithm is
one implementation of public key cryptography.
How do you choose an encryption algorithm?
• No inherent mathematical weakness: Algorithm
survived extensive public review and assume that the
brute force approach is the only efficient attack.
• Key length: A 128-bit key makes a brute force attack
impractical with current technology.
• Key is easy to change and to manage: Frequent key
change makes encryption more secure.
• Cost: Many algorithms are royalty-free.
• Permission for export: Strong cryptography products
may not have permission to export.