@c ========================================================================
@c Begin document body here
@c ========================================================================
@c ========================================================================
@c Conceptual Overview
@c ========================================================================
@node Conceptual Overview
@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{Nodes} 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. @code{Channels} 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 full of obstructions 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. If the peripheral card implemented some networking function,
theys were called Network Interface Cards, or @emph{NICs}. Today most
computers come with the network interface 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{Nodes} in the simulation via @code{Channels}. Just as in a real
computer, a @code{Node} may be connected to more than one @code{Channel} via
multiple @code{NetDevices}.
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{Nodes}, @code{NetDevices} and
@code{Channels}.
Since connecting @code{NetDevices} to @code{Nodes}, @code{NetDevices}
to @code{Channels}, 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, it may take many distinct
@command{ns-3} core operations to create a NetDevice, add a MAC address,
install that net device on a @code{Node}, configure the node's protocol stack,
and then connect the @code{NetDevice} to a @code{Channel}. Even more
operations would be required to connect multiple devices onto multipoint
channels and then to connect individual networks together into internetworks.
We provide topology helper objects that combine those many distinct operations
into an easy to use model for your convenience.
@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, so go ahead and open
@code{first.cc} in your favorite editor.
@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} or shown on the project web page
@uref{http://www.nsnam.org/codingstyle.html,,here}.
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 and contributors have done so with various amounts of
grumbling. 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 licensed 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 above the GPL
text and an author listed below.
@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.
@verbatim
#include "ns3/core-module.h"
#include "ns3/simulator-module.h"
#include "ns3/node-module.h"
#include "ns3/helper-module.h"
@end verbatim
To help our high-level script 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 will recursively load 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 a number of 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} include files is placed in a directory called
@code{ns3} (under the build directory) during the build process to help avoid
include file name collisions. 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, of course, following this tutorial religiously, you will
already have done a
@verbatim
./waf -d debug configure
@end verbatim
in order 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 of these files and find that they
do include all of the public include files in their 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 ``APIs
(Doxygen)'' in the navigation bar. If you select this link, you will be
taken to our documentation page.
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, expand 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 at the top of the page you will see the entry
for @code{NS_LOG_COMPONENT_DEFINE}. Before jumping in, it would probably be
good to look for the ``Detailed Description'' of the logging module to get a
feel for the overall operation. You can either scroll down or select the
``More...'' link under the collaboration diagram to do this.
Once you have a general idea of what is going on, go ahead and take a 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 (script).
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 have seen
that there are a number of levels of logging verbosity/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 during the simulation.
Now we will get directly to the business of creating a topology and running
a simulation. We 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 and 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.SetDeviceAttribute ("DataRate", StringValue ("5Mbps"));
pointToPoint.SetChannelAttribute ("Delay", StringValue ("2ms"));
@end verbatim
The first line
@verbatim
PointToPointHelper pointToPoint;
@end verbatim
instantiates a @code{PointToPointHelper} object on the stack. From a
high-level perspective the next line,
@verbatim
pointToPoint.SetDeviceAttribute ("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} you will find a
``Delay'' attribute associated with the @code{PointToPointChannel}. The
final line,
@verbatim
pointToPoint.SetChannelAttribute ("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
subsequently 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{PointToPointChannel} objects
between them. Just as we used the @code{NodeContainer} topology helper object
to create the @code{Nodes} 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 debug 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.
@verbatim
UdpEchoServerHelper echoServer (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. One of our
conventions is to place required attributes in the helper constructor. In this
case, the helper can't do anything useful unless it is provided with a port
number that the client also knows about. Rather than just picking one and
hoping it all works out, we require the port number as a parameter to the
constructor. The constructor, in turn, simply does a @code{SetAttribute}
with the passed value. You can, if desired, set the ``Port'' attribute to
another value later.
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 holds pointers to all of the applications (one in this case
since we passed a @code{NodeContainer} containing one node) created by the
helper.
Applications require a time to ``start'' generating traffic and may take an
optional time to ``stop.'' We provide both. These times are set using the
@code{ApplicationContainer} methods @code{Start} and @code{Stop}. These
methods take @code{Time} parameters. In this case, we use an explicit C++
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 (the application stop event) to be executed at ten seconds,
the simulation will last at least ten seconds.
@subsubsection UdpEchoClientHelper
The echo client application is set up in a method substantially similar to
that for the server. There is an underlying @code{UdpEchoClientApplication}
that is managed by an @code{UdpEchoClientHelper}.
@verbatim
UdpEchoClientHelper echoClient (interfaces.GetAddress (1), 9);
echoClient.SetAttribute ("MaxPackets", UintegerValue (1));
echoClient.SetAttribute ("Interval", TimeValue (Seconds (1.)));
echoClient.SetAttribute ("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 five different attributes. The
first two attributes are set during construction of the
@code{UdpEchoClientHelper}. We pass parameters that are used (internally to
the helper) to set the ``RemoteAddress'' and ``RemotePort'' attributes in
accordance with our convention to make required attributes parameters in the
helper constructors.
Recall that we used an @code{Ipv4InterfaceContainer} to keep track of the IP
addresses we assigned to our devices. The zeroth interface in the
@code{interfaces} container is going to coorespond 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 first line of code (from above), we
are creating the helper and telling it so set the remote address of the client
to be the IP address assigned to the node on which the server resides. We
also tell it to arrange to send packets to port nine.
The ``MaxPackets'' attribute tells the client the maximum number of packets
we allow it to send during the simulation. The ``Interval'' attribute tells
the client how long to wait between packets, and the ``PacketSize'' attribute
tells the client how large its packet payloads should be. With this
particular 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 application 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
that will be automatically scheduled behind the scenes 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-dev > cp examples/first.cc scratch/
@end verbatim
and then build it using waf,
@verbatim
~/repos/ns-3-dev > ./waf
Entering directory `/home/craigdo/repos/ns-3-dev/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-dev >
@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-dev > ./waf --run scratch/first
Entering directory `/home/craigdo/repos/ns-3-dev/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-dev >
@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.
@c ========================================================================
@c Browsing ns-3
@c ========================================================================
@node Ns-3 Source Code
@section Ns-3 Source Code
Now that you have used some of the @command{ns-3} helpers you may want to
have a look at some of the source code that implements that functionality.
The most recent code can be browsed on our web server at the following link:
@uref{http://code.nsnam.org/?sort=lastchange}. If you click on the bold
repository names on the left of the page, you will see @emph{changelogs} for
these repositories, and links to the @emph{manifest}. From the manifest
links, one can browse the source tree.
The top-level directory for one of our @emph{repositories} will look
something like:
@verbatim
drwxr-xr-x [up]
drwxr-xr-x doc manifest
drwxr-xr-x examples manifest
drwxr-xr-x ns3 manifest
drwxr-xr-x regression manifest
drwxr-xr-x samples manifest
drwxr-xr-x scratch manifest
drwxr-xr-x src manifest
drwxr-xr-x tutorial manifest
drwxr-xr-x utils manifest
-rw-r--r-- 135 .hgignore file | revisions | annotate
-rw-r--r-- 891 .hgtags file | revisions | annotate
-rw-r--r-- 441 AUTHORS file | revisions | annotate
-rw-r--r-- 17987 LICENSE file | revisions | annotate
-rw-r--r-- 4948 README file | revisions | annotate
-rw-r--r-- 4917 RELEASE_NOTES file | revisions | annotate
-rw-r--r-- 7 VERSION file | revisions | annotate
-rwxr-xr-x 99143 waf file | revisions | annotate
-rwxr-xr-x 28 waf.bat file | revisions | annotate
-rw-r--r-- 30584 wscript file | revisions | annotate
@end verbatim
The source code is mainly in the @code{src} directory. You can view source
code by clicking on the @code{manifest} link to the right of the directory
name. If you click on the @code{manifest} link to the right of the src
directory you will find a subdirectory. If you click on the @code{manifest}
link next to the @code{core} subdirectory in under @code{src}, you will find
a list of files. The first file you will find is @code{assert.h}. If you
click on the @code{file} link, you will be sent to the source file for
@code{assert.h}.
Our example scripts are in the @code{examples} directory. The source code for
the helpers we have used in this chapter can be found in the
@code{src/helpers} directory.