TCP/IP, UDP & Multicasting Through Java's Socket
TCP/IP, UDP & Multicasting Through Java's Socket
TCP/IP, UDP & Multicasting Through Java's Socket
Socket
Part I - Basic Networking through Java
Introduction
Since birth of network programming, it has been error-prone, difficult, and complex. The
programmer had to know many details about the network and sometimes even the hardware. You
usually needed to understand the various layers of the networking protocol, and there were a lot
to different functions in each networking library concerned with connecting, packing, unpacking
blocks of information, handshaking, etc. It was a difficult task. However, the concept of
networking is not so difficult. You want to get some information from that machine over there
and move it to this machine here, or vice-versa. Its quite similar to reading and writing files,
except that the files exist on the remote machine.
One of Java's great strengths is painless networking. As much as possible, the underlying details
of networking have been abstracted away. The programming model we use is that of the file
programming model. In addition, Java's built-in multithreading is exceptionally handy when
dealing with another networking issue: handling multiple connections at once. The java.net
package includes classes that enable the programmer to easily pass data across networks.
Networking Concepts
The primary function of the TCP/IP is to provide a point to point communication mechanism.
One process on one machine communicates with the another process on another machine or
within the same machine. This communication appears as two streams of data. One stream
carries data from one process to the other, while the other carries data in the other direction. Each
process can read the data that have been written by the other, and in normal conditions, the data
received are the same, and in the same order, as when they are sent.
In order to tell one machine from another machine and to make sure that you are connected with
the machine you want, there must be some way of uniquely identifying machines on a network.
Early networks were satisfied to provide unique names for machines within the local network.
However, Java works within the Internet, which requires a way to uniquely identify a machine
from all the others in the world. This is accomplished with the IP ( Internet Protocol ) address, a
32 bit number.
In addition to the machine addresses provided by the Internet Protocol part of the network
system, TCP/IP has a mechanism for identifying individual processes on a machine, analogous to
an office block. The building has phone number , but each room inside is also identified by an
extension number. When a call arrives at the building, it must be connected to the correct room
for handling . Payment requests go to accounts payable, orders to sales, and so forth. In the
TCP/IP system, the extension numbers are called ports, and they are represented by a 16-bit
binary number. To communicate with the correct part of a particular computer, the sending
machine must know both the machine address and the port number to which the message should
be sent. Many common services have a dedicated port. Because some ports are reserved for
common services, the programmer cannot use any port. Ports numbered under 1024 are often
referred to as reserved ports, many of which are reserved for a specific program. It is important
that you only attempt to use ports over number 1024.
Connection-oriented programming
Connectionless Programming
Connection-oriented Networking
The client and server have a communication link that is open and active from the time the
application is executed until it is closed. Using Internet jargon, the Transmission control protocol
os a connection oriented protocol. It is reliable connection - packets are guaranteed to arrive in
the order they are sent.
Connection-less Networking
The this type each instance that packets are sent, they are transmitted individually. No link to the
receiver is maintained after the packets arrive. The Internet equivalent is the User Datagram
Protocol (UDP). Connectionless communication is faster but not reliable. Datagrams are used to
implement a connectionless protocol, such as UDP.
Port Number
Service
21
FTP
23
Telnet
25
SMTP(mail)
80
HTTP(Web)
119
NNTP(News)
Client-Server Programming
The most common model of network programming is referred to as client-server programming.
The concept is simple: A client machine makes a request for information or sends a command to
a server; in return, the server passes back the data or results of the command. Most often, the
server only responds to clients; it does not initiate communication.
So the job of the server is to listen for a connection, and that's performed by the special server
object that we create. The job of the client is to try to make a connection to the server, and this is
performed by the special client object we create. Once the connection is made, you will see that
at the server and client ends, the connection is magically just turned into the IO Stream object,
and from then onwards you can treat the connection as if you were reading and writing into the
file.
Socket Terminology
The socket is a software abstraction used to represent the "terminals" of a connection between
two machines or processes. For a given connection , there's a socket on each machine, and you
can imagine a hypothetical "cable" running between the two machines with each end of the
"cable" plugged into the socket. Of course, the physical hardware and cabling between machines
is completely unknown.
In Java, we need to create a socket to make the connection to the other machine. Then you can
get an InputStream and OutputStream from the socket in order to be able to treat the connection
as an IOStream object. There are two stream based socket classes in the java.net package. They
are java.net.ServerSocket that a server uses to listen for incoming connections and a
java.net.Socket that a client uses in order to initiate a connection. Once a client makes a Socket
connection, the ServerSocket returns a corresponding server side socket through which direct
communications will take place.
When we create a ServerSocket, you give it only a port number. You don't have to give it an IP
address because it's already on the machine it represents. When you create a Socket, however,
you must give both the IP address and the port number where you're trying to connect.
Socket Classes
Socket
ServerSocket
DatagramSocket
MulticastSocket( explained in the Part II of this paper )
Before discussing the constructors and methods of Socket and ServerSocket, the class
InetAddress must be mentioned. An InetAddress represents the actual number, not the name or
IP address of a computer. The name Hari.calsoftlabs.com is never used by your program; instead
it uses the corresponding address, 199.2.24.246. The InetAddress class has no constructors,
instead it has some methods that returns the InetAddress address.
Socket
Socket object is the Java representation of a TCP connection. when a socket is created, a
connection is opened to the specified destination.
Constructors:The Socket provides the programmer with four constructors. The address of the
server may be specified as a string or an InetAddress, and the port number on the host to connect
to. In each case, an optional Boolean parameter implements a connectionless socket if set to
false.
Methods:The two most important methods are getInputStream() and getOutputStream(), which
return stream objects that can be used to communicate through the socket. A close() method is
provided to tell the underlying operating system to terminate the connection. Methods are also
provided to retrieve information about the connection to the local host and remote port numbers
and an integers representing the remote host.
ServerSocket
Constructors:The ServerSocket provides two constructors. Both take argument the local port
number to listen for connection requests. A constructor is provided that also takes the maximum
time to wait for a connection as a second argument.
Methods:The most important method is accept(). It returns a Socket that is connected to the
client. The close() method tells the operating system to stop listening for requests on the socket.
Methods to retrieve the host name, the socket is listening on and the port number being listened
to are also provided.
Datagrams
Datagrams are used to implement a connectionless protocol, such as UDP. Two classes are used
to implement datagrams in Java:
java.net.DatagramPacket
java.net.DatagramSocket
DatagramPacket is the actual packet of information, an array of bytes, that is transmitted over the
network. DatagramSocket is a socket that sends and receives DatagramPackets across the
network. You can think of the DatagramPacket as a letter and a DatagramSocket as the mailbox
that the mailcarrier uses to pick up and drop off your letters.
DatagramPacket
The DatagramPacket class provides the programmer with two constructors. The first is used for
DatgramPackets that receive information. This constructor needs to be provided with an array to
store the data and the amount of data to receive. The second is used to create DatagramPackets
that send data. The constructor requires the same information, plus the destination address and
the port number.
Methods: There are four methods in this class - allowing the data, datagram length, and
addressing (InetAddress) and port number information for the packet to be extracted.
DatagramSocket
The DatgramSocket represents a connectionless datagram socket. This class works with the
DatagramPacket class to provide for communication using the UDP protocol. It provides two
constructors, the programmer can specify a port to use or allow the system to randomly use one.
Methods: The two most important methods are - send() and receive(). Each takes as an argument
an appropriately constructed DatagramPacket. In the case of the send() method, the data
contained in the packet is sent to the specified host and the port. The receive() method will block
the execution until a packet is received by the underlying socket, at which time the data will be
copied into the packet provided.
Error handling
Error handling is done by the class called SocketException which extends IOException class.
This exception is thrown when there is a problem using socket. (i.e. ) error in the underlying
protocol, such as a TCP error.
One possible cause is that the local port you are using for is already in use. Another cause is that
the user cannot bind to that particular port. Because, on most operating systems, port numbers
less than 1,024 cannot be used by the programmer except the super user. This is the security
measure, because most well known services reside on the ports in this range.
All the new exceptions extends the SocketException class: They are as follows:
BindException:The local port is in use, or the requested bind address couldn't be assigned
locally.
ConnectException:This exception is raised when a connection is refused at the remote host (i.e.,
no process is listening on that port).
The client sends the message " Hi from client " and the server response will be " Hello from
Server ".
Server Program
import java.net.*;
import java.lang.*;
import java.io.*;
try
{
// Initialising the ServerSocket
sersock = new ServerSocket(PORT);
try
{
sock = sersock.accept();
sock.close();
}
catch(SocketException se)
{
System.out.println("Server Socket
problem "+se.getMessage());
}
catch(Exception e)
{
System.out.println("Couldn't start "
+ e.getMessage()) ;
}
} // main
} // Server class
Client Program
import java.lang.*;
import java.io.*;
import java.net.*;
import java.net.InetAddress;
class client
{
Public static void main(String args[])
{
Socket sock=null;
DataInputStream dis=null;
PrintStream ps=null;
System.out.println(" Trying to connect");
try
{
// to get the ip address of the
server by the name
InetAddress ip =InetAddress.getByName
("Hari.calsoftlabs.com");
}
0catch(SocketException e)
{
System.out.println("SocketException " + e);
}
catch(IOException e)
{
System.out.println("IOException " + e);
}
finally
{
try
{
sock.close();
}
catch(IOException ie)
{
System.out.println(" Close Error :" +
ie.getMessage());
}
} // finally
} // main
} // Class Client
After you've successfully compiled the server and the client programs, you run them. You have
to run the server program first. Just use the Java interpreter and specify the Server class name.
Once the server has started, you can run the client program. After the client sends a request and
receives a response from the server, you should see output similar to this :
On client side:
Hello from Server
On Server side:
Hi from client
A datagram is an independent, self-contained message sent over the network whose arrival,
arrival time, and content are not guaranteed.
The java.net package contains two classes to help you write Java programs that use datagrams to
send and receive packets over the network: DatagramSocket, DatagramPacket, and
MulticastSocket An application can send and receive DatagramPackets through a
DatagramSocket. In addition, DatagramPackets can be broadcast to multiple recipients all
listening to a MulticastSocket.
The following source code demonstrates a slightly more complex server that uses datagrams
instead of sockets. After a connection is made, It echoes back whatever is sent by the client
instead of simply hanging up the connection. It is Called as echo Server.
Server Program
import java.net.*;
import java.io.*;
DatagramPacket packet;
DatagramSocket socket;
byte[] data; // For data to be
Sent in packets
int clientPort;
InetAddress address;
String str;
for(;;){
data = new byte[packetSize];
socket.receive(packet);
}catch(IOException ie)
{
System.out.println(" Could not Receive
:"+ie.getMessage());
System.exit(0);
}
address = packet.getAddress();
clientPort = packet.getPort();
try
{
// sends packet
socket.send(packet);
}
catch(IOException ex)
{
System.out.println("Could not Send
"+ex.getMessage());
System.exit(0);
}
} // for loop
} // main
} // class EchoServer
Client Program
import java.net.*;
import java.io.*;
if(args.length != 2)
{
System.out.println("Usage Error :
Java EchoClient < Server name> < Message>");
System.exit(0);
}
try
{
// sends the packet
socket.send(packet);
}
catch(IOException ie)
{
System.out.println("Could not Send :"+ie.getMessage());
System.exit(0);
}
try
{
// Receives the packet from the server
socket.receive(packet);
}
catch(IOException iee)
{
System.out.println("Could not receive :
"+iee.getMessage() );
System.exit(0);
}
} // Class EchoClient
The client side and the server side networking code looks actually very similar.This is true with
many applications that use datagrams because the java.net.DatagramSocket class is used to both
send and receive DatagramPackets.
Suppose server running on the machine named Hari.calsoftlabs.com, whereas the client running
on the xxxx.info.com. As you can see at the end of the example the server is running waiting for
the another connection, while the execution of the client has halted.
Server Side :
Java EchoServer
Output:
Message:Hello
From :xxxx.info.com
Client Side :
Output:
Conclusion
In this paper introduction to full spectrum of techniques in java networking are explained. We
have seen techniques for programming clients and servers using both TCP and UDP services. If
you have previous experience of network programming, you will probably agree that java makes
network programming much easier than do most other languages. Thus, one of Java's great
strengths is painless networking. As much as possible, the underlying details of networking have
been abstracted away.
Most high-level network protocols only provide a unicast transmission service. That is, nodes of
the network only have the ability to send to one other node at a time. All transmission with a
unicast service is inherently point-to-point. If a node wants to send the same information to many
destinations using a unicast transport service, it must perform a replicated unicast, and send N
copies of the data to each destination in turn.
A better way to transmit data from one source to many destinations is to provide a multicast
transport service. With a multicast transport service, a single node can send data to many
destinations by making just a single call on the transport service:
For those applications which involve a single node sending to many recipients, a multicast
facility is clearly a more natural programming paradigm than unicast. When a multicast service
is implemented over such a network, there is a huge improvement in performance. If the
hardware supports multicast, A packet which is destined for N recipients can be sent as just a
single packet!
Multicast Groups
The notion of "group" is essential to the concept of multicasting. By definition a multicast
message is sent from a source to a group of destination hosts. In IP multicasting, multicast
groups have an ID called multicast group ID. Whenever a multicast message is sent out, a
multicast group ID specifies the destination group. These group ID's are essentially a set of IP
addresses called "Class D" explained in the following section. Therefore, if a host (a process in a
host) wants to receive a multicast message sent to a particular group, it needs to somehow listens
to all messages sent to that particular group. Different Ipv4 addresses are explained in the
following section.
Unicast addresses are used for transmitting a message to a single destination node.
Broadcast addresses are used when a message is supposed to be transmitted to all nodes in a
subnetwork.
For delivering a message to a group of destination nodes which are not necessarily in the same
subnetwork,multicast addresses are used.
Class A, B, and C IP addresses are used for unicast messages, whereas as class D IP address,
those in the range 224.0.0.1 to 239.255.255.255, inclusive, and by a standard UDP port number
are used for multicast messages.
Bit no. 0 1 2 3 4 5 6 7 8 16 24 31
Class D addresses are identified by a one in bit 0,1 and 2 and a zero in bit 3 of the address. This
means that 6.25% of all available IP addresses are of this class.
The range of Class D addresses are in dotted decimal notation from 224.h.h.h.h to 239.h.h.h,
where h is a number from 0 to 255. Address 224.0.0.0 is reserved and can not be used, while
address 224.0.0.1 is used to address all hosts that take part in IP multicasting.
Class D addresses are used for multicasting and does not have a network part and hosts part. IP
multicasting makes it possible to send IP datagrams to a group of hosts, which may be spread
across many networks.
Take for example, a TV station where TTLs would be the station's signal area -- the limitation of
how far the information can travel. As the packet moved around the company's internal network,
its TTL would be notched down every time it passed through an router. When the packet's TTL
reached 0, the packet would die and not be passed further. Generally multicast with long TTLs --
perhaps 200 - to guarantee that the information will reach around the world
Java's Support
Java makes the programmer easier by implementing classes in java.net package that facilitates
our need i.e multicasting. java.net includes a class called MulticastSocket. This kind of socket is
used on the client-side to listen for packets that the server broadcasts to multiple clients. The
multicast datagram socket class is useful for sending and receiving IP multicast packets. Thus,
java.net.MulticastSocket. and java.net.DatagramPacket together used to implement multicasting
in java and makes programming easier.
Constructors:
public MulticastSocket() throws IOException
This allows us to Create a multicast socket.
Methods:Methods other than the joinGroup(addr) and leaveGroup(addr), this class also provides
functions, that set and gets the time-to-live for multicast packets sent out on the socket. The TTL
sets the IP time-to-live for DatagramPackets sent to a MulticastGroup, which specifies how
many "hops" that the packet will be forwarded on the network before it expires. They are
setTTl(ttl) and getTTl(ttl). The parameter 'ttl' is an unsigned 8-bit quantity.
The Following source code is an simple example for broadcasting. It broadcast date and time to
its multiple clients. ( from java 1.1 developers handbook )
Server Program
import java.net.*;
import java.io.*;
import java.util.*;
public class BroadcastServer
{
public static final int PORT = 1200;
public static void main(String args[])
throws Exception{ MulticastSocket socket;
DatagramPacket packet; InetAddress address;
address = InetAddress.getByName();
socket = new MulticastSocket();
// join a Multicast group and send the
group salutations socket.joinGroup(address);
byte[] data = null;
for(;;)
{
Thread.sleep(1000);
System.out.println("Sending ");
String str = (new Date()).toString();
data = str.getBytes();
packet = new DatagramPacket
(data,str.length(),address,PORT);
// Sends the packet socket.send(packet);
}
//
for
}
// main
} // class BroadcastServer
Client program
import java.net.*;
import java.io.*;
MulticastSocket socket;
DatagramPacket packet;
InetAddress address;
address = InetAddress.getByName(args[0]);
socket = new MulticastSocket(BroadcastServer.PORT);
socket.joinGroup(address);
byte[] data = null;
packet = new DatagramPacket(data,data.length);
for(;;)
{
// receive the packets
socket.receive(packet);
String str = new String(packet.getData());
System.out.println(" Time signal received from"+
packet.getAddress() + " Time is : " +str);
} // for
} // main
} // class Broadcast
First you will need to ensure that a name is associated with a suitable multi-cast address. This
requires access to - and understanding of your system's configuration. Add a suitable name, such
as BroadTest and assign it the address between 224.0.0.1 to 239.255.255.255.
You will see a message "Sending " appear at one-second intervals. This suggests that a server is
running.
You should now see the date and time printed at regular intervals on the client machine
Advantages of Multicasting
In many cases the multicasting capability is desirable because of some following advantages.
The first major advantage of using multicasting is the decrease of the network load. There are
many applications like stock ticker applications which are required to transmit packets to
hundreds of stations. The packets sent to these stations share a group of links on their paths to
their destinations. Since multicasting requires the transmission of only a single packet by the
source and replicates this packet only if it is necessary multicast transmission can conserve the so
much needed network bandwidth.
Another place where multicasting can be very helpful is in resource discovery. There are many
applications in which a host needs to find out whether a certain type of service is available or
not. Using multicast messages and sending the query to those hosts which are potentially capable
of providing this service would be of great help. Although some applications use multicast
messages to transmit a packet to a group of hosts residing on the same network, there is no
reason to impose this limitation. The scope of multicast packets can be limited by using the time-
to-live (TTL) field of these packets.
Another important feature of multicasting is its support for datacasting applications. In recent
years, multimedia transmission has become more and more popular. The audio and video signals
are captured, compresses and transmitted to a group of receiving stations. Instead of using a set
of point-to-point connections between the participating nodes, multicasting can be used for
distribution of the multimedia data to the receivers. In real world stations may join or leave an
audio-cast or a video-cast at any time.
Today, the MBONE is a critical piece of the technology that's needed to make multiple person
data voice, and video conferencing on the Internet -- in fact, sharing any digital information --
cheap and convenient. Unfortunately, the majority of the routers on the Internet today don't know
how to handle multicasting. Most routers are set up to move traditional Internet Protocol (IP)
unicast packets -- information that has a single, specific destination. Most routers today are
unicast routers: They are designed to move information from a specific place to another specific
place. However, routers that include multicasting capabilities are becoming more common.
So in order to facilitate multicasting MBONE network was created. It's a "virtual network" - a
network that runs on top of the Internet and it allows multicast packets to traverse the Net. The
MBONE is called a virtual network because it shares the same physical media -- wires, routers
and other equipment -- as the Internet.
The MBONE allows multicast packets to travel through routers that are set up to handle only
unicast traffic. Software that utilizes the MBONE hides the multicast packets in traditional
unicast packets so that unicast routers can handle the information.The scheme of moving
multicast packets by putting them in regular unicast packets is called tunneling. In the future,
most commercial routers will support multicasting, eliminating the headaches of tunneling
information through unicast routers.
Machines (workstations or routers) that are equipped to support multicast IP are called mrouters
(multicast routers). Mrouters are either commercial routers that can handle multicasting or (more
commonly) dedicated workstations running special software that works in conjunction with
standard routers. Multicasting is a network routing facility -- a method of sending packets to
more than one site at a time. The MBONE is a loose confederation of sites that currently
implement IP multicasting.
MBONE Advantages
In the future, the MBONE may make it possible for you to do the following:
Conclusion
In this paper, we first reviewed why and for what applications multicasting is important. Then,
the essential concept of group membership was discussed. Here a sample program in java for
broadcasting shows the easiness to implement such a complex muticasting. Thus java with it's
MulticastSocket class and methods provides the easier way to implement the multicasting by
abstracting the underlying details of networking. It also introduces the concept of MBONE and
it's advantages.