Transcript Networking - Rabie A. Ramadan

Advanced Programming
Rabie A. Ramadan
[email protected]
Network Programming
An Overview
2
Computer Network

A computer network is an interconnected collection of
autonomous computers.
Network topology

A topology is a way of “laying out” the network.
Topologies can be either physical or logical.

Physical topologies describe how the cables are run.

Logical topologies describe how the network messages
travel
Network topology (cont.)

Bus (can be both logical and physical)

Star (physical only)

Ring (can be both logical and physical)

Mesh (can be both logical and physical)
Network topology (cont.)
Bus


A bus is the simplest physical topology. It consists of a single
cable that runs to every workstation
This topology uses the least amount of cabling, but also covers
the shortest amount of distance.
Network topology (cont.)

It is difficult to add a workstation

Have to completely reroute the cable and possibly run
two additional lengths of it.

If any one of the cables breaks, the entire network is
disrupted. Therefore, it is very expensive to maintain.
Network topology (cont.)
Star Topology


A physical star topology branches each network device off a
central device called a hub, making it very easy to add a new
workstation.
Also, if any workstation goes down it does not affect the entire
network.
Network topology (cont.)

Star topologies are more expensive to install than
bus networks,
• There are several more cables that need to be installed,
plus the cost of the hubs that are needed.
Network topology (cont.)
Ring

Each computer connects to two other computers,
joining them in a circle creating a unidirectional path
where messages move workstation to workstation.
Network topology (cont.)

The ring makes it difficult to add new computers.

Unlike a star topology network, the ring topology
network will go down if one entity is removed from the
ring.

Physical ring topology systems don’t exist much
anymore, mainly because the hardware involved was
fairly expensive and the fault tolerance was very low.
Network topology (cont.)
Mesh


The mesh topology is the simplest logical topology in
terms of data flow, but it is the most complex in terms of
physical design.
In this physical topology, each device is connected to
every other device
Network topology (cont.)

The physical mesh topology is very expensive to
install and maintain.

Cables must be run from each device to every
other device. The advantage you gain from it is its
high fault tolerance.

There will always be a way of getting the data
from source to destination.
Network topology (cont.)

Advantages and Disadvantages of Network Topologies
Topology
Advantages
Disadvantages
Bus
Cheap. Easy to install.
Difficult to reconfigure.
Break in bus disables
entire network.
Star
Cheap. Easy to install.
Easy to reconfigure.
Fault tolerant.
More expensive than bus.
Ring
Efficient. Easy to install.
Reconfiguration difficult.
Very expensive.
Mesh
Simplest. Most fault tolerant.
Reconfiguration extremely difficult.
Extremely expensive.
Very complex.
Computer Network

A network includes:
• Special purpose hardware devices that:
• Interconnect transmission media
• Control transmission of data
• Run protocol software
• Protocol software that:
• Encodes and formats data
• Detects and corrects problems encountered during
transmission
Network Hardware
Repeaters



Repeaters are very simple devices.
Allow a cabling system to extend beyond its maximum allowed
length by amplifying the network voltages so they travel
farther.
Repeaters are nothing more than amplifiers and, as such, are
very inexpensive.
Network Hardware
Hubs

Hubs are devices used to link several computers together.

They repeat any signal that comes in on one port and copy it to
the other ports (a process that is also called broadcasting).

There are two types of hubs: active and passive.

•
•
Passive hubs simply connect all ports together electrically and are usually not
powered.
Active hubs use electronics to amplify and clean up the signal before it is
broadcast to the other ports.
In the category of active hubs, there is also a class called
“intelligent” hubs, which are hubs that can be remotely
managed on the network.
Network Hardware
Hubs
Network Hardware
Bridges


They join similar topologies and are used to divide network segments.
For example, with 200 people on one Ethernet segment, the performance
will be mediocre, because of the design of Ethernet and the number of
workstations that are fighting to transmit. If you divide the segment into
two segments of 100 workstations each, the traffic will be much lower
on either side and performance will increase.

They are more intelligent than repeaters but are unable to move data
across multiple networks simultaneously.

Unlike repeaters, bridges can filter out noise.

The main disadvantage to bridges is that they can’t connect dissimilar
network types or perform intelligent path selection. For that function,
you would need a router.
Network Hardware
Bridges
Network Hardware
Routers




Routers are highly intelligent devices that connect multiple
network types and determine the best path for sending data.
The advantage of using a router over a bridge is that routers
can determine the best path that data can take to get to its
destination.
Like bridges, they can segment large networks and can filter
out noise.
However, they are slower than bridges because they are more
intelligent devices;
Network Hardware
Switch

Just like a hub

A network switch conserves network bandwidth and offers generally better
performance than a hub.

A vital difference between a hub and a switch is that all the nodes connected
to a hub share the bandwidth among themselves, while a device connected to a
switch port has the full bandwidth all to itself.

For example, if 10 nodes are communicating using a hub on a 10-Mbps
network, then each node may only get a portion of the 10 Mbps if other nodes
on the hub want to communicate as well. .

But with a switch, each node could possibly communicate at the full 10 Mbps.
Network Hardware
Switch
Addressing and Routing

Address: byte-string that identifies a node

Routing: process of forwarding messages to the
destination node based on its address

Types of addresses
• usually unique
• unicast: node-specific
• broadcast: all nodes on the network
• multicast: some subset of nodes on the network
IP Addresses and Classes
IP Addresses and Classes
IP Addresses and Classes
IP Addresses and Classes
28
Subnet Mask
29
Network Architecture

A network architecture is a set of layers and protocols used
to reduce network design complexity.

The TCP/IP Protocol Suite (also called the Internet
Architecture) is an important example of a network
architecture.

The OSI (Open Systems Interconnection) 7-Layer
Reference Model [ISO,1984] is a guide that specifies what
each layer should do, but not how each layer is implemented.
ISO/OSI Reference Model
ISO 7-Layer Reference Model
End host
End host
Application
Application
Presentation
Presentation
Session
Session
Transport
Transport
Network
Network
Network
Network
Data link
Data link
Data link
Data link
Physical
Physical
Physical
Physical
One or more nodes
within the network
TCP/IP Model
33
Internet Model
34
Protocols

A protocol is a set of rules of communication. Protocols are the
building blocks of a network architecture.

Term “protocol” is overloaded
• specification of peer-to-peer interface
• module that implements this interface
Network Programming

A network allows arbitrary applications to
communicate.

However, a network programmer doesn’t need to
know the details of all lower-level network
technologies.

Network facilities are accessed through an
Application Programming Interface (API);
e.g., a Service Interface.
Basic Paradigm for
Communication

Most network applications can be divided into two
pieces: a client and a server.

A Web browser (a client) communicate with a Web
server.

A Telnet client that we use to log in to a remote host.

A user who needs access to data located at remote
server.
Basic Paradigm for
Communication



Establish contact (connection).
Exchange information (bi-directional).
Terminate contact.
Client-Server Paradigm

Server waits for client to request a connection.

Client contacts server to establish a connection.

Client sends request.

Server sends reply.

Client and/or server terminate connection.
Two types of Communication

Connection-oriented
• Setup the link before communication.
• Similar to the phone call. We need the phone number
and receiver.

Connectionless
• No link needed to be set up before communication.
• Similar to send a letter. We need the address and
receiver.
TCP and UDP


TCP (Transmission Control Protocol) is a connection-oriented
protocol.
UDP (User Datagram Protocol) is connectionless (UDP)
protocol.
Ports
Identifying the ultimate destination



IP addresses identify hosts
Host has many applications
Ports (16-bit identifier)
Application
Port
WWW
80
E-mail
25
Telnet
23
192.18.22.13
Sockets

A socket is defined as an endpoint for communication.

Concatenation of IP address and port

A socket pair (local IP address, local port, foreign IP address,
foreign port) uniquely identifies a communication.

The socket 161.25.19.8:1625 refers to port 1625 on host
161.25.19.8
Sockets
44
Sockets and Ports
socket
any port
agreed port
socket
message
client
server
other ports
Internet address = 138.37.94.248
Internet address = 138.37.88.249
TCP Socket
Bind() binds the socket to the specified
address. The address parameter specifies
the local component of the address, e.g. IP
address and UDP/TCP port
UDP Socket
Types of Sockets

A stream socket uses TCP for
connection-oriented
communication. It is also called a
TCP socket.

A datagram socket uses UDP for
connectionless communication. It
is also called a UDP socket.

A raw socket uses IP directly. It
is used in such applications as in
ICMP protocol
ICMP: Internet Control Message
Protocol

ICMP: Internet Control
Message Protocol
• ICMP handles error and
control information between
gateways and hosts.
• The messages are transmitted
using IP datagrams.
• They are normally generated
and processed by the TCP/IP
software itself, not user
processes.
UNIX TCP Communication


Normally, a server would first listen and accept a connection
and then fork a new process to communicate with the client.
It uses the listen operation to listen for connections on a socket.
 int listen (int sockfd, int backlog) :
- Sockfd a socket descriptor created by the socket call.
- The backlog parameter defines the maximum length the
queue of pending connections may grow to.
UNIX TCP Communication

The server uses the accept system call to
accept connection requested by a client.

After a connection has been established,
both processes may then use the write
(send) and read (recv) operations to send
and receive messages.
Establishing a TCP Connection
Passive Open



Server process does a ‘passive’ open on a port
It waits for a client to connect
At this stage there is no Internet network traffic
Establishing a TCP Connection
Active Open

client process
usually on a
different
machine

performs an
‘active’ open on
the port
Example - Programming Server



Initialization:
•
•
•
socket - create socket
bind – bind socket to the local address
listen - associate socket with incoming requests
Loop:
•
•
•
accept - accept incoming connection
recv - receive message from client
send - send message to client
Termination:
•
close - close connection socket
Example - Programming Client



Initialization:
•
•
•
gethostbyname - look up server
socket - create socket
connect - connect to server port
Transmission:
•
•
send – send message to server
recv - receive message from server
Termination:
•
close - close socket
Windows Sockets




Similar to UNIX, only the interface is different
We call it Windows Socket or WinSock
Fits above the transport layer
It is not a Protocol -- Located in a dynamic linked library (DLL).
56
Windows Sockets
•
Windows places a single copy of socket functions and TCP/IP
code in memory.
• All applications share these functions/code
57
Steps in creating Clients in Java
58
Steps in creating Clients in Java
59
60
61
62
63
64
65
66
67
68
69
70
References




http://www.sockets.com/winsock.htm
http://tangentsoft.net/wskfaq/
Core Web Programming book
Chapter 17. Network Programming
Assignment is posted