The network administrators of UNITEN will benefit from this simulation process since the simulation can realize what is happening in the network and how much traffic and traffic types the network can support for maximum extent.
To simulate the campus network effectively, the simulation software called simulator is needed. There are a number of simulators existing in the market which include academic editions, commercial edition and open source. The study and comparison of available network simulators will be done and the simulation process 4.
Page 4 of 68 will be conducted by using the most suitable simulator for campus network simulation process. The Gantt chart is attached in appendix Appendix A. The network simulators help the network designers and administrators to check whether the network can perform as it is expected to do so in real time or not.
To conduct network functionality test by using network simulators can reduce a lot of time and money, and also providing facilities to deploy the networks which are equal or similar to real time scenario [4].
Page 6 of 68 2. The recent emulation involves the special purpose and stand-alone network emulators which support packet delays and drops. The recent emulators looks likes the routers to the end station which include PCs, Laptops and mobile phones. Those emulators were developed by developers where they implement the systems as kernel drop-in modules which intercept Network Layer packet forwarding path.
Network emulators are limited in capability and the maximum capability of emulator is simple packet manipulation. The simple packet manipulation does not provide interference from simulated cross traffic. The emulators do not support general simulation capabilities which the network simulators do provide. The model has the similarity to the system which the model represents but the model has higher simplicity than the system.
The purpose of modeling is to predict the effect of the changes made to the system. A model should have the similarity with the real system as much as possible and even should inherit the salient features of the system.
On top of that, the model should not be very complex so that it is easy to study and conduct experiment with it. A good model has reality and simplicity. The complexity of the model will have to increase iteratively during the experiment performing with the model. One of the important issues in modeling is the model validity.
To validate the model, the experiment with the same input must be conducted on both model and the real system 7. Page 7 of 68 which the model represents.
If the output from the model and the real system are the same, the model is valid. The aim of network planning and design is to ensure that new architecture or design meet the requirement of the customers. All users have to pay for the license or the specific module package which they want to use for their specific project requirements.
One of the typical example is OPNET which has two different license type namely; commercial and academic license. For the users who want to use 8. Page 8 of 68 OPNET for their commercial projects, they have to pay for the license and for the academic researchers who wants to do their research and development in network area, they will have to register with their institutional email in order to receive the academic license. Commercial simulators have advantages and disadvantages.
The advantages is relatively more that the disadvantages such as learning resource availability, the software capability, ease of use, complete and up to date documentations since they are perfectly maintained by group of people who are specialized for software and documents maintenance in that company.
The open source network simulator has also advantages and disadvantages. Disadvantage is relatively more than advantages particularly in documentation. Generally, to do maintenance and documentation for open source software, there are not enough specialized people. This documentation problem can become very serious if there is new version of software comes with many new features and functions, then the users will be very difficult to trace or understand the previous source codes without appropriate documentations.
Lack of enough documentation support and lack of version control support can lead to serious matter such as the applicability and life time of the network simulators. The detail discussions of mentioned network simulated will be conducting in following sections. Page 9 of 68 2. In general, a minimum function of network simulator include the network simulator should enable users to represent a network topology, defining scenarios, specifying the nodes including node type and function on the network, the links between those nodes and the traffic generation capability among those nodes.
The more complicated network simulator allows users to specify everything about the protocols used to process network traffic i.
Network Layer Protocols. Another aspect of simplicity and complexity of network simulator is graphical user interface GUI. Network simulators which have GUI enabled, allow users to easily visualize their works in the simulated environment. Some of the simulators may base on text- based user interface which provide lesser visual or intuitive interface.
Some of them may be programming- oriented and can provide programming framework that allows users to create an application which generate network traffic in network testing environment. Page 10 of 68 2. OPNET Modeler has flexibility that can be used to study communication networks, devices, protocols, and applications. The graphical interface can be used to create network topologies and entities from Application Layer to Physical Layer.
Object Oriented Programming technique is used to create the mapping from the graphical design to the implementation of real systems. The Figure 2. The users can see all the topology configurations and simulation results which can be presented very intuitively and visually. The simulation parameter such as amount of traffic generated, amount of time for traffic generation or simulation can be adjusted to repeat the simulation for different scenarios by using GUI.
Page 11 of 68 Figure 2. Hierarchical structure is used to organize the network which is unique to simulate hierarchical network architectures. The programming tools are also required to simulate the networks since the tools provide techniques to define the state of transition machine, network model, and process modules.
Figure 2. Page 12 of 68 Figure 2. The GUI interface and Programing tools are very useful to help users to build systems as complex as they want. Page 13 of 68 2. For modeling, it provide graphical user interface which is very intuitive. Users can create all kinds of models of protocols. For analysis, the simulation results and data can be analyzed and displayed using the graphs and also generated web report. User friendly graphs, charts, statistics, and even animations can be generated by OPNET analysis module for convenience of users.
Page 14 of 68 2. The original NS was discrete event network simulator targeted for network research and development. The first version was developed in The current second version NS2 is widely used in academic research and has a lot of packages contributed by non-profit organizations and groups. The biggest reason to use these two programming languages is due to the internal characteristics and features of these two programming languages.
It is not easy to modify and assemble different components and to change different parameters without very visual and easy to use descriptive languages. On top of that, for efficiency reason, NS2 implemented control path and data path separately. The event scheduler and basic network components object reside in data path.
Therefore, the Page 15 of 68 combinations of these two programming languages prove to be very effective and efficient. To initiate the event scheduler, set up topology, and tell packet traffic source when to start and stop sending packet according to event scheduler, OTcl script is used. The scenario can be changed easily by programming the OTcl script. If the user wants to create the new network object, the user can either write the new object or assemble a compound object from the existing object library, and plug the data path through the object.
The plugging makes NS2 very powerful. Page 16 of 68 2. NS3 is designed to replace the current NS2. Page 17 of 68 o Tracing architecture: NS3 is developing a tracing and statistics gathering framework trying to enable customization of the output without rewriting or rebuilding the simulation core. Page 18 of 68 2. The modules are assembled into larger models by using high level programming language.
Page 19 of 68 2. Page 20 of 68 2. NS2 has larger number of available models, realistic mobility models, powerful and flexible scripting and simulation setup, large user community and ongoing development.
NS2 needs to be recompilation every time if there is a change in the user code. NS2 provides an easy traffic and movement pattern by including an efficient energy model. Real system is too complex and it is complicated infrastructure to model.
NS2 provides a set of randomized mobility model and there are several projects to bring advanced mobility models to simulators. The mixture of compilation and interpretation made it difficult to analyse and to understand the code. Complex scenarios can be easily tested by using NS2. Including a lot of nodes in NS2 may slow down the simulation process. NS2 is popular for its modularity. Page 21 of 68 2. NS3 has high modularity than its ancestor NS2. NS3 still suffers from lack of credibility. NS3 is intended to replicate the successful mode of NS2 in which a lot of different organizations contributed to the models and components based on the framework of NS2.
NS3 support for ported code which should make model validation easier and more credible. NS3 is much more flexible than NS2. NS3 requires active maintainers to respond to the user questions and bug reports, to help test and validate the system 5. Wide range of use in both optimization and expansion of the existing networks. Page 22 of 68 2. OPNET leverage three different simulation technologies to efficiently trade off simulations detail and speed 1.
Complex GUI operation 2. OPNET has fast discrete event simulation engine. OPNET does not allow much number of nodes within a single connected device. OPNET has customizable wireless modelling. OPET result accuracy is by the sampling resolution.
OPNET simulation is inefficient if nothing happens for long periods. Page 23 of 68 2. Page 24 of 68 2. Page 26 of 68 2. Each of the events occurs in particular time interval and this event marks a change of system state. Between two consecutive events, there is no changes in the system is assumed.
Therefore, the simulation can directly jump from one event to another event. The state trajectory overtime can be represented mathematically by a step function [17].
The values of step function change with respect to discrete events. The clock skips to next event start time as the simulation proceeds. This is sometimes called pending event set since there are lists of events which are pending to be simulated if the previous simulated event is not yet finished. An event is denoted by the time at which it occurs and type. When the events are instantaneous, the activities that extend over time are modeled as Page 27 of 68 sequences of events.
Single threaded simulation engines which are based on instantaneous events have only one current event. Multi-threaded simulation engine and simulation engines which support interval-based event modal may have multiple current events. To accomplish the random number generation tasks, one or more Pseudorandom number generator [18]. The use of pseudo random numbers are opposite from the use of true random numbers. The benefit is if the simulation needs to re run again, the simulation will run again with the exactly same behavior.
Therefore, the simulation designer has to decide when the simulation should end. Page 28 of 68 2. For example if a person will take to finish certain task with 10 minutes, the person object in the simulation process would take 10 minutes to finish the task as well. Real time simulation is dedicated and commonly used for computer gaming. Real time simulators are used extensively in many of the engineering fields of study such as statistical power grid protection tests, aircraft design and simulation, motor drive controller design methods and space robot integration.
Continuous simulation allows prediction of rocket trajectories, hydrogen bomb dynamics, electric circuit simulation and robotics. Continuous simulations are based on sets of differential equation which defines peculiarity of the state variables, the environment factors to speak, of a system.
In the continuous simulation method, the conceptual model must be constructed by using set of differential equations. To develop a conceptual model, there are two approaches namely deductive approach and inductive approach.
The deductive approach states that the behavior of system arises from physical laws that can be applied while the inductive approach states that the behavior of the system arises from observed behavior of an actual example.
Page 29 of 68 2. Network Topology can be subdivided into physical network topology and logical network topology [24]. Page 30 of 68 2. The top level node may be a computer, a switch, or just a common connection point.
Messages received by the top level node such as Hub can further be broadcasted to all subordinate nodes or if the top level node such as switch is smart enough to send only to the desired subordinate node.
Inter node messaging delays are reduced with this configuration. Failure in the connection between top level node and subordinate node or failure in the subordinate node will not disrupt the entire network.
This is an important advantage of Star Network Topology. Star Network topology is commonly used in a LAN [25] which span in larger geometric area.
One of the disadvantages of Star Network Topology is that it needs more cabling than any other network topologies. Page 31 of 68 2. This topology is preferred if the traffic between node is large. Every node in this type of network have multiple to another destination node.
The probability of single point network failure is greatly minimized with the Mesh Network Topology. The major advantages of Mesh Network Topology is that source node determine the best route from sender to destination based on connectivity, speed, pending node tasks. A disadvantage of Mesh Network Topology is that it incur large cost when implementation. In Full Mesh, each node being directly connected to every other node in the network while in Partial Mesh, having some nodes in the network being connected indirectly to the others in the network.
Page 32 of 68 Figure 2. Logical topology of the network can be dynamically reconfigured when selecting network equipment such as router. Page 33 of 68 2. RIP sends out the complete routing table to all active interfaces in every 30 seconds. RIP use hop count as the cost to determine the best route to remote network. This hop count is limited to 15 by default and meaning that the remote located at 16th hop count is not reachable.
RIP is only suitable for small networks and if the network is too large, RIP becoming inefficient due to large numbers of router installed in the network. RIP version support only class-full routing, this means that all devices in the network must use same subnet mask. RIP version 1 does not send subnet mask information when it sends the routing table update.
In RIP version 2, there is something called prefix routing and hence support classless routing by sending subnet mask information in route updates [26].
BGP is routing protocol which operates at the network layer. Routing domain is also known as Autonomous System AS because different administrative authorities control their own respective domains. The current internet is the network of interconnected routing domains where BGP version 4 is the de facto routing protocol [26].
Page 34 of 68 2. It was developed by organization called Xerox and the specification was called Aloha net. Ethernet LAN typically used coaxial cable or special grades of twisted pair wires. Wireless LAN also uses Ethernet. The devices which are connected to cable are in shared media. Ethernet even provide even higher level backbone support at Mbps. The redundancy and availability is critical in the core layer. Normally, high Page 35 of 68 performance routers are used as core layer devices since there is network address translation between intranet and outside internet.
After receiving the data from access layer devices, distribution layer switches transmit the data to core network layer. High performance switches are used in distribution since the load of data is relatively bigger than access layer and to make sure the network has redundancy and high availability. End user devices include PCs, Laptops. Printers, Mobile phones and Tablets. FTP is usable via a terminal by the user but is designed mainly for use by programs. Page 36 of 68 2. HTTP is stateless application layer protocol for distributed, collaborative and hypertext information systems.
HTTP is popular in retrieving web related documents [30]. The motivations of internet telephony are very low cost, demand for multimedia communication and demand for integration of voice and data network. The message must contain a message header and message text formatted [32]. Page 37 of 68 2.
POP is one of the methods of accessing email. POP3 does not maintain folder hierarchy in server side so that user may have difficulty to retrieve email if user has to access from different computer rather than usual one [33]. IMAP is one of the methods of accessing email as well.
IMAP provides server-side storage and manipulation of email. IMAP maintain email folder hierarchy in server side [33]. TCP is connection oriented since TCP implement the mechanism to setup and tear down a full duplex connection between end points. The data sent over network via TCP is guaranteed that the receiver will receive correctly and completely since TCP has mechanism for error free and in order delivery of data.
From the Dock, click the downloaded connect. You can find the downloaded connect. A dialog box will display with details of the Certificate of the computer you are connecting to. Select Continue. This method requires access to the Android Play Store. You can find out if your Chromebook supports Android apps by clicking here. If you already have the Chrome OS Webstore version of Global Protect, please uninstall it by right-clicking or alt-clicking the application and selecting Uninstall.
A connect. Optional step: You may wish to change the default program used to open. Select RD Client from the options provided. Now you can double click any.
When using for the first time, you may see a dialog box will ask you to accept agreements. Select Accept. Request to the microphone, bluetooth and camera will be required. Click Continue on each prompt that appears. Open your Files app, locate the connect file and tap to open. The file should open in Microsoft Remote Desktop connection. EL Student Remote Access Lab. In order to run specialist software applications such as matlab, opnet, sage etc from a PC in one of the two Linux labs A2.
As detailed in the diagram below, connect from your personal computer, through the gateway server, onto one of the machines in the specialist Linux labs. The diagram gives an example of the lab machines — see a complete list of lab machines you can connect to. The lab machines are dual-boot, and automatically boot into Linux overnight, however, if the machine is switched off, or being used during the day by Windows users, you will not be able to connect to that specific machine and you may need to try an alternative machine.
Open a terminal and connect to the gateway server using the following command, substituting username with your login credentials. Accept the fingerprint when prompted, followed by your password. You will now be able to launch your application from the remote PC, such as Opnet, which will display on your own machine. Type in the name of the application you wish to run in the terminal window example: [ opnet ] or [ firefox ] or [ matlab ] or [ sage ].
In order to connect into the Linux system using a Windows based machine you will need to have some extra software which will allow remote access to a Linux X11 window graphics environment. We recommend using the successfully tested graphical application called Xming, along with access software such as Putty. Connect to the gateway entry server Select Session from the menu on the left, and type ssh. If this is the first time you are connecting to the gateway server from this machine, you will be prompted to accept a security key fingerprint.
Select Yes to this prompt. You will now be able to ssh on to a machine in one of the Linux labs A2. Install the Xquartz program for X11 display xquartz is the official X server software for mac.
Connect to the gateway server using the following command, substituting username with your login credentials. Connect to the lab machine, again substituting username with your login credentials, and hostname with the name of a machine in either A2. Go to Service Desk.
This site uses cookies. Click here to view our cookie policy message. Windows 10 Guide Remote access to lab computers. Access and download software. Introducing Windows Do your research remotely. Everything you need to connect remotely to a lab computer from your device. Getting started When using a computer remotely: You will need to be connected to the VPN You are only able to connect to PCs that are available— the status of a computer will become unavailable as soon as the computer has been logged into You will need to have a reliable internet connection You have to log into the computer with your University account Some software is only available in some locations Standard software such as Microsoft Office and web browsers is available on all computers The resolution of the monitor may be different to what you are using to connect with Computer connections are reserved for 3 minutes, if you do not connect during this time, the connection file will time-out and you may need to select a different computer to connect to If your session is inactive for 30 minutes you will be disconnected You can reuse an Remote Desktop Connection RDP connection if the computer remains available — please check before attempting to reuse a connection.
How to connect. Close all. A Connect dialog window will appear, click Download to download the connection file.
0コメント