unison/doc/tutorial/conceptual-overview.texi

@c ========================================================================
@c Begin document body here
@c ========================================================================

@c ========================================================================
@c Conceptual Overview
@c ========================================================================
@node Conceptual Overveiw
@chapter Conceptual Overview

@menu
* Key Abstractions::
* A First ns-3 script
@end menu

The first thing we need to do before actually starting to look at or write
@command{ns-3} code is to explain a few core concepts and abstractions in the
system.  Much of this may appear transparently obvious to some, but we
recommend taking the time to read through this section just to ensure you
are starting on a firm foundation.

@node Key Abstractions
@section Key Abstractions

In this section, we'll review some terms that are commonly used in
networking, but have a specific meaning in @command{ns-3}.

@subsection Node
@cindex Node
In Internet jargon, a computing device that connects to a network is called
a @emph{host} or sometimes an @emph{end system}.  Because @command{ns-3} is a 
@emph{network} simulator, not specifically an @emph{Internet} simulator, we 
intentionally do not use the term host since it is closely associated with
the Internet and its protocols.  Instead, we use a more generic term also
used by other simulators that originates in Graph Theory --- the @emph{node}.

@cindex class Node
In @command{ns-3} the basic computing device abstraction is called the 
node.  This abstraction is represented in C++ by the class @code{Node}.  The 
@code{Node} class provides methods for managing the representations of 
computing devices in simulations.

You should think of a @code{Node} as a computer to which you will add 
functionality.  One adds things like applications, protocol stacks and
peripheral cards with their associated drivers to enable the computer to do
useful work.  We use the same basic model in @command{ns-3}.

@subsection Application
@cindex Application
Typically, computer software is divided into two broad classes.  @emph{System
Software} organizes various computer resources such as memory, processor
cycles, disk, network, etc., according to some computing model.  System
software usually does not use those resources to complete tasks that directly
benefit a user.  A user would typically run an @emph{application} that acquires
and uses the resources controlled by the system software to accomplish some
goal.  

@cindex system call
Often, the line of separation between system and application software is made
at the privilege level change that happens in operating system traps.
In @command{ns-3} there is no real concept of operating system and especially
no concept of privilege levels or system calls.  We do, however, have the
idea of an application.  Just as software applications run on computers to
perform tasks in the ``real world,'' @command{ns-3} applications run on
@command{ns-3} @code{Node}s to drive simulations in the simulated world.

@cindex class Application
In @command{ns-3} the basic abstraction for a user program that generates some
activity to be simulated is the application.  This abstraction is represented 
in C++ by the class @code{Application}.  The @code{Application} class provides 
methods for managing the representations of our version of user-level 
applications in simulations.  Developers are expected to specialize the
@code{Application} class in the object-oriented programming sense to create new
applications.  In this tutorial, we will use specializations of class 
@code{Application} called @code{UdpEchoClientApplication} and 
@code{UdpEchoServerApplication}.  As you might expect, these applications 
compose a client/server application set used to generate and echo simulated 
network packets 

@subsection Channel
@cindex Channel

In the real world, one can connect a computer to a network.  Often the media
over which data flows in these netowrks are called @emph{channels}.  When
you connect your Ethernet cable to the plug in the wall, you are connecting 
your computer to an Ethernet communication channel.  In the simulated world
of @command{ns-3} one connects a @code{Node} to an object representing a
communication channel.  Here the basic communication subnetwork abstraction 
is called the channel and is represented in C++ by the class @code{Channel}.  

The @code{Channel} class provides methods for managing communication 
subnetwork objects and connecting nodes to them.  They may also be specialized
by developers in the object oriented programming sense.  A @code{Channel}
specialization may model something as simple as a wire.  The specialized 
@code{Channel} can also model things as complicated as a large Ethernet
switch, or three-dimensional space in the case of wireless networks.

We will use specialized versions of the @code{Channel} called
@code{CsmaChannel}, @code{PointToPointChannel} and @code{WifiChannel} in this
tutorial.  The @code{CsmaChannel}, for example, models a version of a 
communication subnetwork that implements a @emph{carrier sense multiple 
access} communication medium.  This gives us Ethernet-like functionality.  

@subsection Net Device
@cindex NetDevice
@cindex Ethernet
It used to be the case that if you wanted to connect a computers to a network,
you had to buy a specific kind of network cable and a hardware device called
(in PC terminology) a @emph{peripheral card} that needed to be installed in
your computer.  These cards were called Network Interface Cards, or 
@emph{NIC}s.  Today most computers come with the network controller hardware
built in and users don't see these building blocks.

A NIC will not work without a software driver to control the hardware.  In 
Unix (or Linux), a piece of peripheral hardware is classified as a 
@emph{device}.  Devices are controlled using @emph{device drivers}, and network
devices (NICs) are controlled using @emph{network device drivers}
collectively known as @emph{net devices}.  In Unix and Linux you refer
to these net devices by names such as @emph{eth0}.

In @command{ns-3} the @emph{net device} abstraction covers both the software 
driver and the simulated hardware.  A net device is ``installed'' in a 
@code{Node} in order to enable the @code{Node} to communicate with other 
@code{Node}s in the simulation via @code{Channel}s.  Just as in a real
computer, a @code{Node} may be connected to more than one @code{Channel} via
multiple @code{NetDevice}s.

The net device abstraction is represented in C++ by the class @code{NetDevice}.
The @code{NetDevice} class provides methods for managing connections to 
@code{Node} and @code{Channel} objects; and may be specialized by developers
in the object-oriented programming sense.  We will use the several specialized
versions of the @code{NetDevice} called @code{CsmaNetDevice},
@code{PointToPointNetDevice}, and @code{WifiNetDevice} in this tutorial.
Just as an Ethernet NIC is designed to work with an Ethernet network, the
@code{CsmaNetDevice} is designed to work with a @code{CsmaChannel}; the
@code{PointToPointNetDevice} is designed to work with a 
@code{PointToPointChannel} and a @code{WifiNetNevice} is designed to work with
a @code{WifiChannel}.

@subsection Topology Helpers
@cindex helper
@cindex topology
@cindex topology helper
In a real network, you will find host computers with added (or built-in)
NICs.  In @command{ns-3} we would say that you will find @code{Nodes} with 
attached @code{NetDevices}.  In a large simulated network you will need to 
arrange many connections between @code{Node}s, @code{NetDevice}s and 
@code{Channel}s.

Since connecting @code{NetDevice}s to @code{Node}s, @code{NetDevice}s
to @code{Channel}s, assigning IP addresses,  etc., are such common tasks
in @command{ns-3}, we provide what we call @emph{topology helpers} to make 
this as easy as possible.  For example, will take several distinct 
@command{ns-3} core operations to create a NetDevice, add a MAC address, 
connect the net device to a @code{Node}, configure the protocol stack, and 
then connect the @code{NetDevice} to a @code{Channel}.  More operations would
be required to connect multiple devices onto multipoint channels and then to 
connect networks together into internetworks.  We use topology helper objects
to combine those many distinct operations into an easy to use model.

@c ========================================================================
@c A First ns-3 script
@c ========================================================================
@node A First ns-3 Script
@section A First ns-3 script
@cindex first script
If you downloaded the system as was suggested above, you will have a release
of @command{ns-3} in a directory called @code{repos} under your home 
directory.  Change into that release directory, and you should find a 
directory structure something like the following:

@verbatim
  AUTHORS  examples/  README         samples/  utils/   waf.bat*
  build/   LICENSE    regression/    scratch/  VERSION  wscript
  doc/     ns3/       RELEASE_NOTES  src/      waf*
@end verbatim

@cindex first.cc
Change into the examples directory.  You should see a file named 
@code{first.cc} located there.  This is a script that will create a simple
point-to-point link between two nodes and echo a single packet between the
nodes.  Let's take a look at that script line by line.

@subsection Boilerplate
The first line in the file is an emacs mode line.  This tells emacs about the
formatting conventions (coding style) we use in our source code.  

@verbatim
  /* -*- Mode:C++; c-file-style:''gnu''; indent-tabs-mode:nil; -*- */
@end verbatim

This is always a somewhat controversial subject, so we might as well get it
out of the way immediately.  The @code{ns-3} project, like most large 
projects, has adopted a coding style to which all contributed code must 
adhere.  If you want to contribute your code to the project, you will 
eventually have to conform to the @command{ns-3} coding standard as described 
in the file @code{doc/codingstd.txt}.  We recommend that you, well, just get 
used to the look and feel of @code{ns-3} code and adopt this standard whenever
you are working with our code.  All of the development team have done so.
The emacs mode line above makes it easier to get the formatting correct if you
use the emacs editor.

The @command{ns-3} simulator is licentsed using the GNU General Public 
License.  You will see the appropriate GNU legalese at the head of every file 
in the @command{ns-3} distribution.  Often you will see a copyright notice for
one of the institutions involved in the @code{ns-3} project and an author
listed.

@verbatim
  /*
   * This program is free software; you can redistribute it and/or modify
   * it under the terms of the GNU General Public License version 2 as
   * published by the Free Software Foundation;
   *
   * This program is distributed in the hope that it will be useful,
   * but WITHOUT ANY WARRANTY; without even the implied warranty of
   * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   * GNU General Public License for more details.
   *
   * You should have received a copy of the GNU General Public License
   * along with this program; if not, write to the Free Software
   * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
   */
@end verbatim

@subsection Module Includes
The code proper starts with a number of include statements.  To help our high
level users deal with the large number of include files present in the system,
we group includes according to relatively large modules.  We provide a single
include file that, in turn, includes all of the include files used in each
module.  Rather than having to look up exactly what header you need, and
possibly have to get dependencies right, we give you the ability to load a 
group of files at a large granularity.  This is not the most efficient approach
but it certainly makes writing scripts much easier.  Each of the 
@command{ns-3} files is placed in a directory called @code{ns3} (under the 
build directory) to help avoid include file name collisions.

@verbatim
  #include "ns3/core-module.h"
  #include "ns3/simulator-module.h"
  #include "ns3/node-module.h"
  #include "ns3/helper-module.h"
@end verbatim

The @code{ns3/core-module.h} file corresponds to the ns-3 module you will find
in the directory @code{src/core} in your downloaded release distribution.  If 
you list this directory you will find a large number of header files.  When
you do a build, Waf will place public header files in an @code{ns3} directory
under the appropriate @code{build/debug} or @code{build/optimized} directory
depending on your configuration.  Waf will also automatically generate a module
include file to load all of the public header files.  Since you are following
this tutorial religiously, you will already have done a

@verbatim
  ./waf -d debug configure
@end verbatim

to configure the project to perform debug builds.  You will also have done a

@verbatim
  ./waf
@end verbatim

to build the project.  So now if you look in the directory 
@code{build/debug/ns-3} you will find the four module include files shown 
above.  You can take a look at the contents to find that these files will
include all of the public includes in the respective modules.

@subsection Ns3 Namespace
The next line in the @code{first.cc} script is a namespace declaration.

@verbatim
  using namespace ns3;
@end verbatim

The @command{ns-3} project is implemented in a C++ namespace called 
@code{ns3}.  This groups all @command{ns-3}-related declarations in a scope
outside the global namespace, which we hope will help with integration with 
other code.  The C++ @code{using} statement introduces the @code{ns-3}
namespace into the current (global) declarative region.  This is a fancy way
of saying that after this declaration, you will not have to type @code{ns3::}
scope resolution operator before all of the @code{ns-3} code in order to use
it.  If you are unfamiliar with namespaces, please consult almost any C++ 
tutorial and compare the @code{ns3} namespace and usage here with instances of
the @code{std} namespace and the @code{using namespace std;} statements you 
will often find in discussions of @code{cout} and streams.

@subsection Logging
The next line of the script is the following,

@verbatim
  NS_LOG_COMPONENT_DEFINE ("FirstScriptExample");
@end verbatim

We will use this statement as a convenient place to talk about our Doxygen
documentation system.  If you look at the project web site, 
@uref{http://www.nsnam.org,,ns-3 project}, you will find a link to ``Other
Documentation'' in the navigation bar.  If you select this link, you will be
taken to our documentation page.  You will find links to our development
@code{ns-3-dev} documentation as well as that for our latest release.  If you
select the @code{HTML} link you will be taken to the Doxygen documentation for
that version.  

Along the left side, you will find a graphical representation of the structure
of the documentation.  A good place to start is the @code{NS-3 Modules} 
``book.''  If you expand @code{Modules} you will see a list of @command{ns-3}
module documentation.  The concept of module here ties directly into the 
module include files discussed above.  It turns out that the @command{ns-3}
logging subsystem is part of the @code{core} module, so go ahead and expand 
that documentation node.  Now, open the @code{Debugging} book and then select 
the @code{Logging} page.

You should now be looking at the Doxygen documentation for the Logging module.
In the list of @code{#define}s you will see @code{NS_LOG_COMPONENT_DEFINE}.  
It would probably be good to look for the ``Detailed Description'' of the 
logging module now to get a feel for the overall operation and then look at
the specific @code{NS_LOG_COMPONENT_DEFINE} documentation.  I won't duplicate
the documentation here, but to summarize, this line declares a logging 
component called @code{FirstScriptExample} that allows you to enable and 
disable console message logging by reference to the name.

@subsection Main Function
The next lines of the script you will find are:

@verbatim
    int
  main (int argc, char *argv[])
  {
@end verbatim

This is just the declaration of the main function of your program.  Just as in
any C++ program, you need to define a main function that will be the first 
function run.  There is nothing at all special here.  Your @command{ns-3}
script is just a C++ program.

The next two lines of the script are used to enable two logging components that
are built into the Echo Client and Echo Server applications:

@verbatim
    LogComponentEnable("UdpEchoClientApplication", LOG_LEVEL_INFO);
    LogComponentEnable("UdpEchoServerApplication", LOG_LEVEL_INFO);
@end verbatim

If you have read over the Logging component documentation you will see that
there are a number of levels of detail that you can enable on each component.  
These two lines of code enable debug logging at the INFO level for echo 
clients and servers.  This will result in the application printing out 
messages as packets are sent and received.

Now we will get directly to the business of creating a topology and running 
a simulation.  We will use the topology helper objects to make this job as
easy as possible.

@subsection Topology Helpers
@subsubsection NodeContainer
The next two lines of code in our script will actually create the 
@command{ns-3} @code{Node} objects that will represent the computers in the 
simulation.  

@verbatim
    NodeContainer nodes;
    nodes.Create (2);
@end verbatim

Let's find the documentation for the @code{NodeContainer} class before we
continue.  Another way to get into the documentation for a given class is via 
the @code{Classes} tab in the Doxygen pages.  If you still have the Doxygen 
handy, just scroll up to the top of the page and select the @code{Classes} 
tab.  You should see a new set of tabs appear, one of which is 
@code{Class List}.  Under that tab you will see a list of all of the 
@command{ns-3} classes.  Scroll down, looking for @code{ns3::NodeContainer}.
When you find the class, go ahead and select it to go to the documentation for
the class.

You may recall that one of our key abstractions is the @code{Node}.  This
represents a computer to which we are going to add things like protocol stacks,
applications and peripheral cards.  The @code{NodeContainer} topology helper
provides a convenient way to create, manage and access any @code{Node} objects
that we create in order to run a simulation.  The first line above just 
declares a NodeContainer which we call @code{nodes}.  The second line calls the
@code{Create} method on the @code{nodes} object that asks the container to 
create two nodes.  As described in the Doxygen, the container calls down into
the @command{ns-3} system proper to create two @code{Node} objects and stores
pointers to those objects internally.

The nodes as they stand in the script do nothing.  The next step in 
constructing a topology is to connect our nodes together into a network.
The simplest form of network we support is a single point-to-point link 
between two nodes.  We'll construct one of those links here.

@subsubsection PointToPointHelper
We are constructing a point to point link, and, in a pattern which will become
quite familiar to you, we use a topology helper object to do the low-level
work required to put the link together.  Recall that two of our key 
abstractions are the @code{NetDevice} and the @code{Channel}.  In the real
world, these terms correspond roughly to peripheral cards and network cables.  
Typically these two things are intimately tied together and one cannot expect
to interchange, for example, Ethernet devices and wireless channels.  Our 
Topology Helpers follow this intimate coupling and therefore you will use a
single @code{PointToPointHelper} to configure and connect @command{ns-3}
@code{PointToPointNetDevice} and @code{PointToPointChannel} objects in this 
script.

The next three lines in the script are,

@verbatim
    PointToPointHelper pointToPoint;
    pointToPoint.SetDeviceParameter ("DataRate", StringValue ("5Mbps"));
    pointToPoint.SetChannelParameter ("Delay", StringValue ("2ms"));
@end verbatim

The first line 

@verbatim
    PointToPointHelper pointToPoint;
@end verbatim

creates a @code{PointToPointHelper} object on the stack.  From a high-level 
perspective the next line,

@verbatim
    pointToPoint.SetDeviceParameter ("DataRate", StringValue ("5Mbps"));
@end verbatim

tells the @code{PointToPointHelper} object to use the value ``5mbps''
(five megabits per second) as the ``DataRate'' when it creates a 
@code{PointToPointNetDevice} object.

From a more detailed perspective, the string ``DataRate'' corresponds
to what we call an @code{Attribute} of the @code{PointToPointNetDevice}.
If you look at the Doxygen for class @code{ns3::PointToPointNetDevice} and 
find the documentation for the @code{GetTypeId} method, you will find a list
of  @code{Attributes} defined for the device.  Among these is the ``DataRate''
attribute.  Most user-visible @command{ns-3} objects have similar lists of 
attributes.  We use this mechanism to easily configure simulations without
recompiling as you will see in a following section.

Similar to the ``DataRate'' on the @code{PointToPointNetDevice} we find a
``Delay'' attribute associated with the @code{PointToPointChannel}.  The 
final line,

@verbatim
    pointToPoint.SetChannelParameter ("Delay", StringValue ("2ms"));
@end verbatim

tells the @code{PointToPointHelper} to use the value ``2ms'' (two milliseconds)
as the value of the transmission delay of every point to point channel it 
creates.

@subsubsection NetDeviceContainer
At this point in the script, we have a @code{NodeContainer} that contains
two nodes.  We have a @code{PointToPointHelper} that is primed and ready to 
make @code{PointToPointNetDevices} and wire @code{PoiintToPointChannel} objects
between them.  Just as we used the @code{NodeContainer} topology helper object
to create the @code{Node}s for our simulation, we will ask the 
@code{PointToPointHelper} to do the work involved in creating, configuring and
installing our devices for us.  We will need to have a list of all of the 
NetDevice objects that are created, so we use a NetDeviceContainer to hold 
them just as we used a NodeContainer to hold the nodes we created.  The 
following two lines of code,

@verbatim
    NetDeviceContainer devices;
    devices = pointToPoint.Install (nodes);
@end verbatim

will finish configuring the devices and channel.  The first line declares the 
device container mentioned above and the second does the heavy lifting.  The 
@code{Install} method of the @code{PointToPointHelper} takes a 
@code{NodeContainer} as a parameter.  Internally, a @code{NetDeviceContainer} 
is created.  For each node in the @code{NodeContainer} (there must be exactly 
two for a point-to-point link) a @code{PointToPointNetDevice} is created and 
saved in the device container.  A @code{PointToPointChannel} is created and 
the two @code{PointToPointNetDevices} are attached.  When objects are created
by the @code{PointToPointHelper}, the attributes previously set in the helper
are used to initialize the corresponding attributes in the created objects.

After executing the the @code{pointToPoint.Install (nodes)} call we will have
two nodes, each with an installed point-to-point net device and a 
point-to-point channel between them.  Both devices will be configured to 
transmit data at five megabits per second over the channel which has a two 
millisecond transmission delay.

@subsubsection InternetStackHelper
We now have nodes and devices configured, but we don't have any protocol stacks
installed on our nodes.  The next two lines of code will take care of that.

@verbatim
    InternetStackHelper stack;
    stack.Install (nodes);
@end verbatim

The @code{InternetStackHelper} is a topology helper that is to internet stacks
what the @code{PointToPointHelper} is to point-to-point net devices.  The
@code{Install} method takes a @code{NodeContainer} as a parameter.  When it is
executed, it will install an Internet Stack (TCP, UDP, IP, etc.) on each of
the nodes in the node container.

@subsubsection Ipv4AddressHelper
Next we need to associate the devices on our nodes with IP addresses.  We 
provide a topology helper to manage the allocation of IP addresses.  The only
user-visible API is to set the base IP address and network mask to use when
performing the actual address allocation (which is done at a lower level 
inside the helper).

The next two lines of code in our example script, @code{first.cc},

@verbatim
    Ipv4AddressHelper address;
    address.SetBase ("10.1.1.0", "255.255.255.0");
@end verbatim

declare an address helper object and tell it that it should begin allocating IP
addresses from the network 10.1.1.0 using the mask 255.255.255.0 to define 
the allocatable bits.  By default the addresses allocated will start at one
and increase monotonically, so the first address allocated from this base will
be 10.1.1.1, followed by 10.1.1.2, etc.  The low level @command{ns-3} system
actually remembers all of the IP addresses allocated and will generate a
fatal error if you accidentally cause the same address to be generated twice 
(which is a very hard to find error, by the way).

The next line of code,

@verbatim
    Ipv4InterfaceContainer interfaces = address.Assign (devices);
@end verbatim

performs the actual address assignment.  In @command{ns-3} we make the
association between an IP address and a device using an @code{Ipv4Interface}
object.  Just as we sometimes need a list of net devices created by a helper 
for future reference we sometimes need a list of @code{Ipv4Interface} objects.
The @code{Ipv4InterfaceContainer} provides this functionality.

Now we have a point-to-point network built, with stacks installed and IP 
addresses assigned.  What we need at this point are applications to generate
traffic.

@subsection Applications
Another one of the core abstractions of the ns-3 system is the 
@code{Application}.  In this script we use two specializations of the core
@command{ns-3} class @code{Application} called @code{UdpEchoServerApplication}
and @code{UdpEchoClientApplication}.  Just as we have in our previous 
explanations,  we use helper objects to help configure and manage the 
underlying objects.  Here, we use @code{UdpEchoServerHelper} and
@code{UdpEchoClientHelper} objects to make our lives easier.

@subsubsection UdpEchoServerHelper
The following lines of code in our example script, @code{first.cc}, are used
to set up a UDP echo server application on one of the nodes we have previously
created and connected using a point-to-point link.  

@verbatim
    UdpEchoServerHelper echoServer;
    echoServer.SetPort (9);

    ApplicationContainer serverApps = echoServer.Install (nodes.Get (1));
    serverApps.Start (Seconds (1.0));
    serverApps.Stop (Seconds (10.0));
@end verbatim

The first line of code in the above snippet declares the 
@code{UdpEchoServerHelper}.  As usual, this isn't the application itself, it
is an object used to help us create the actual applications.  The second line
that has the @code{SetPort} call, is used to tell the helper to assign the
value nine to the ``Port'' attribute when creating 
@code{UdpEchoServerApplication} objects.

Similar to many other helper objects, the @code{UdpEchoServerHelper} object 
has an @code{Install} method.  It is the execution of this method that actually
causes the underlying echo server application to be instantiated and attached
to a node.  Interestingly, the @code{Install} method takes a
@code{NodeContainter} as a parameter just as the other @code{Install} methods
we have seen.  This is actually what is passed to the method even though it 
doesn't look so in this case.  There is a C++ @emph{implicit conversion} at
work here.

We now see that @code{echoServer.Install} is going to install a
@code{UdpEchoServerApplication} on the node found at index number one of the
@code{NodeContainer} we used to manage our nodes.  @code{Install} will return
a container that has all of the applications (one in this case since we passed
a @code{NodeContainer} containing one node) made by the helper.  

Applications require a time to ``start'' generating traffic and a time to
``stop.''  These times are set using @code{ApplicationContainer} methods
@code{Start} and @code{Stop}.  These methods take @code{Time} parameters.  
In this case, we use an explicit conversion sequence to take the C++ double
1.0 and convert it to an @command{ns-3} @code{Time} object using a 
@code{Seconds ()} cast.  The two lines,

@verbatim
    serverApps.Start (Seconds (1.0));
    serverApps.Stop (Seconds (10.0));
@end verbatim

will cause the echo server application to @code{Start} (enable itself) at one
second into the simulation and to @code{Stop} (disable itself) at ten seconds
into the simulation.  By virtue of the fact that we have implicilty declared
a simulation event at ten seconds, the simulation will last at least ten 
seconds.

@subsubsection UdpEchoClientHelper

The echo client application is set up in a substantially method similar to the
server.  There is an underlying @code{UdpEchoClientApplication} that is 
managed by an @code{UdpEchoClientHelper}.

@verbatim
    UdpEchoClientHelper echoClient;
    echoClient.SetRemote (interfaces.GetAddress (1), 9);
    echoClient.SetAppAttribute ("MaxPackets", UintegerValue (1));
    echoClient.SetAppAttribute ("Interval", TimeValue (Seconds (1.)));
    echoClient.SetAppAttribute ("PacketSize", UintegerValue (1024));

    ApplicationContainer clientApps = echoClient.Install (nodes.Get (0));
    clientApps.Start (Seconds (2.0));
    clientApps.Stop (Seconds (10.0));
@end verbatim

For the echo client, however, we need to set four different attributes.  The 
first attribute is set using the @code{SetRemote} method.  Recall that
we used an @code{Ipv4InterfaceContainer} to keep track of the IP addresses we
assigned to our devices.  As a result of the allocation, the zeroth interface
in the @code{interfaces} container cooresponds to the IP address of the 
zeroth node in the @code{nodes} container.  The first interface in the 
@code{interfaces} container cooresponds to the IP address of the first node in
the @code{nodes} container.  So, in the following line of code (reproduced 
from above), we are setting the remote address of the client to be the IP 
address assigned to the node on which the server resides, and we tell it to 
send packets to port nine.

@verbatim
    echoClient.SetRemote (interfaces.GetAddress (1), 9);
@end verbatim

The ``MaxPackets'' attribute tells the client the maximum number of packets 
it can send.  The ``Interval'' attribute tells the client how long to wait 
between packets, and the ``PacketSize'' attribute tells the client how large 
its packets should be.  With this combination of attributes, we are telling 
the client to send one 1024-byte packet.

Just as in the case of the echo server, we tell the echo client to @code{Start}
and @code{Stop}, but here we start the client one second after the server is
enabled (at two seconds into the simulation).

@subsection Simulator
What we need to do at this point is to actually run the simulation.  This is 
done using the global function @code{Simulator::Run}.

@verbatim
    Simulator::Run ();
@end verbatim

When we previously called the methods,

@verbatim
    serverApps.Start (Seconds (1.0));
    serverApps.Stop (Seconds (10.0));
    ...
    clientApps.Start (Seconds (2.0));
    clientApps.Stop (Seconds (10.0));
@end verbatim

we actually scheduled events in the simulator at 1.0 seconds, 2.0 seconds and
10.0 seconds.  When @code{Simulator::Run} is called, the ssytem will begin 
looking through the list of scheduled events and executing them.  First it 
will run the event at 1.0 seconds, which will enable the echo server 
application.  Then it will run the event scheduled for t=2.0 seconds which 
will start the echo client application.  The start event implementation in 
the echo client will begin the data transfer phase of the simulation by 
sending a packet to the server.

The act of sending the packet to the server will trigger a chain of events
which will be automatically scheduled and which will perform the mechanics of
the packet echo according to the various timing parameters that we have set 
in the script.

Eventually, since we only send one packet, the chain of events triggered by 
that single client echo request will taper off and the simulation will go 
idle.  Once this happens, the remaining events will be the @code{Stop} events
for the server and the client.  When these events are executed, there are
no further events to process and @code{Simulator::Run} returns.  The simulation
is complete.

All that remains is to clean up.  This is done by calling the global function 
@code{Simulator::Destroy}.  As the helper functions (or low level 
@command{ns-3} code) executed, they arranged it so that hooks were inserted in
the simulator to destroy all of the objects that were created.  You did not 
have to keep track of any of these objects yourself --- all you had to do 
was to call @code{Simulator::Destroy} and exit.  The @command{ns-3} system
took care of the hard part for you.  The remaining lines of our first 
@command{ns-3} script, @code{first.cc}, do just that:

@verbatim
    Simulator::Destroy ();
    return 0;
  }
@end verbatim

@subsection Building Your Script
We have made it trivial to build your simple scripts.  All you have to do is 
to drop your script into the scratch directory and it will automatically be 
built if you run Waf.  Let's try it.  Copy @code{examples/first.cc} into 
the @code{scratch} directory.

@verbatim
  ~/repos/ns-3-tutorial > cp examples/first.cc scratch/
@end verbatim

and then build it using waf,

@verbatim
  ~/repos/ns-3-tutorial > ./waf
  Entering directory `/home/craigdo/repos/ns-3-tutorial/build'
  [432/477] cxx: scratch/first.cc -> build/debug/scratch/first_2.o
  [475/477] cxx_link: build/debug/scratch/first_2.o ...
  Compilation finished successfully
~/repos/ns-3-tutorial >
@end verbatim

You can now run the example (note that if you build your program in the scratch
directory you must run it out of the scratch direcory):

@verbatim
  ~/repos/ns-3-tutorial > ./waf --run scratch/first
  Entering directory `/home/craigdo/repos/ns-3-tutorial/build'
  Compilation finished successfully
  Sent 1024 bytes to 10.1.1.2
  Received 1024 bytes from 10.1.1.1
  Received 1024 bytes from 10.1.1.2
  ~/repos/ns-3-tutorial >
@end verbatim

Here you see that the build system checks to make sure that the file has been
build and then runs it.  You see the logging component on the echo client 
indicate that it has sent one 1024 byte packet to the Echo Server on 
10.1.1.2.  You also see the logging component on the echo server say that
it has received the 1024 bytes from 10.1.1.1.  The echo server silently 
echoes the packet and you see the echo client log that it has received its 
packet back from the server.