DemoNet

MIS Department

Fox School of Business and Management

Temple University

Computer systems networking is one of many basic skills that MIS students must develop in order to compete in the IT industry. Without basic networking skills IT professionals cannot effectively plan, deploy, and support the applications which bring value to their organizations. As such, most MIS programs include one or more courses that focus on networking as part of their required curriculum.

One of the issues that instructors consistently struggle with is how to best present networking concepts in the classroom. While discussing networking concepts, often guided by PowerPoint presentations, can be used to communicate some basic ideas, demonstrations of how the technology works in the real world have a powerful impact on a student’s understanding and retention.

The Perfect Demonstration Network

In the perfect classroom, instructors would have a fully functional network at their disposal for demonstration purposes. In the perfect lab, students would have access to this network to explore how networks really work. The network would be small and simple enough for students to understand but include all of the components that you would find in a production network. This network would include LANs created with switches. It would include backbone networks and WANs created with routers. Students could ping workstations and trace the routes of packets through the network from one location to another. As any production network would have in the real world, this network would have redundant DNS and DHCP servers that provided services to the network. This network could be used to demonstrate all of the most essential networking concepts.

The perfect network would go beyond basic networking technologies and include a variety of servers ranging from SQL servers to IIS servers to web, application, and file/print servers. How applications run across our network could be demonstrated using these servers.

The perfect network would include a comprehensive directory such as Microsoft’s Active Directory (AD). All of the machines in this network would be part of an AD domain. From users and groups to policies that manage everything from software distribution to patch management, the AD could be used to manage the environment. This network would be monitored and managed just like a production network in the real world using tools that are used in production networks.


The perfect network might look something like this:

The MIS department at Temple University’s Fox School of Business has recently adopted a case study that is being utilized in every one of its undergraduate courses. By utilizing the same case study throughout the program, students get to explore the business problem from the complete spectrum of perspectives. This case study revolves around the merger of two fictitious banks, Glenside and Jenkintown banks. While the bulk of the case study focuses on the challenges related to disparate business processes and data, the network diagram shown above depicts the physical network that supports these two organizations.

Glenside has a single location with three LANs; one LAN on each floor of the building and a third LAN in their datacenter. These LANs are connected via a backbone network. Jenkintown has one main location. Much like Glenside, Jenkintown has a small datacenter, a separate LAN for workstations, connected to each other with a backbone network. Jenkintown also has five branch offices; one of these five branch offices is depicted in our network diagram. The same routers that are used to create the backbone networks in Glenside and Jenkintown are also used to create the WAN which connects, Glenside, Jenkintown and the various branch offices.

Fiction Transformed into Reality

In addition to utilizing this case study in every course in the MIS program, the students and faculty of the MIS department have built the network described above. This is a fully functioning network. Everything you see in the network diagram exists and is fully operational. Students can login to AD accounts on workstations in the second floor of the Glenside office and have drives mapped to file servers based on group policy. They can PING workstations over in the branch office and use TRACERT to examine how packets flow from one location to another. They can issue an IPCONFIG command, find their IP address and calculate the valid address on their particular subnet. In addition, they can explore the DHCP servers and see how these servers manage addresses for the entire network. From basic networking to security to systems management, the list of valuable lessons that can be demonstrated with this network go on and on. For a more complete list of demonstrations, see Appendix A.

This network, known as DemoNet, is used on a regular basis in our undergraduate and graduate networking classes to demonstrate a wide variety of networking concepts. Since the SQL servers are fully functioning SQL servers complete with mirrored boot drives, RAID-5 arrays for housing data, and separate mirrored volumes for transaction logs, we are exploring the use of the DemoNet SQL servers in our database classes. DemoNet’s collection of database, application servers, and web servers are also candidates for use in a variety of application development and system integration courses.

Cost Effective Implementation

Universities have not typically created networks like this for demonstration purposes. The time, money, and energy required to implement a complete network like this is prohibitive. The hardware costs alone could easily exceed $100,000. This is far too great of an investment for an individual university with a handful of networking courses to make. In addition, finding the resources with the ability to build and maintain a comprehensive network can be challenging.

As a result, many faculty members all over the world spend time and energy cobbling together individual components of a network for demonstrations. These typically only include the components that the faculty members have the most experience with. This is rarely the most effective use of a faculty member’s time and talents.

The students and faculty in the MIS department at Temple University have utilized an extremely innovative approach to building this network. This network has been built on top of virtual machine technology from VMWare, Inc. This technology enables us to run multiple virtual machines, both workstations and servers, on a single physical machine. VMWare includes their own virtual switching technology which enables us to create LANs. By building servers with multiple virtual network interface cards which are connected to virtual switches and by enabling routing on these servers, routers were created to support the backbone networks and WAN connections. The end result is a fully functional network, complete with redundant components that all run within a single physical system, all at a fraction of the cost of a physical network.

Network In A Box

One of the other benefits of this innovative approach is that, by being built on virtual machine technology, the entire network is nothing more than a collection of files. The entire network can be replicated from one physical system to another with little more than a simple copy-n-paste with a mouse. An entire network can be restored to its pristine state by simply copying files.

The MIS department is currently exploring options for commercializing this environment, making this network available to other universities for use in their networking class. In one scenario that is being explored the MIS department would host a virtual network for a peer university. This network would be accessed via remote desktop across the Internet. In another scenario the MIS department would make available the virtual machines that comprise this network to peer universities who would host the virtual network internally.

Like To Learn More?

If you’d like to learn more about DemoNet and explore opportunities to utilize DemoNet in your classes please contact Mart Doyle (). In addition, if you have ideas for any additional capabilities that you’d like to see added to DemoNet, again, please contact Mart Doyle.

Appendix A – Sample Demonstrations

Demonstrations have been grouped into the following categories:

1.  Networking Demonstrations

  1. LANs and IP addresses – In this demonstration we look at the IP configuration of machines in the datacenter and each floor of the Glenside location. We determine the range of valid IP address for each subnet in the Glenside location. In addition to examining the IP addresses of the hosts, we look at the default gateway for machines on each LAN. We then look at the router in the Glenside location which supports both the backbone network within the Glenside location and connectivity to the Jenkintown location. We match up the IP addresses of each of the network interfaces on the router with the default gateway addresses used on each of the LANs.
  2. TRACERT across the backbone network and the WAN – In this demonstration we use the PING command and the TRACERT command to explore connectivity and how packets are routed throughout the network
  3. DNS tour – In this demonstration we examine the DNS servers in our network. We see examples of both A-records. CNAME records are used. We also see how DHCP clients are registered in DNS. We temporarily shutdown DNS servers and see how the pair of redundant DNS servers continue to provide name services even when one server is down.
  4. DHCP demonstration – In this demonstration we work with both DHCP clients and servers. Workstations request and receive IP addresses from DHCP servers. After reviewing the IP configuration of a DHCP client, we look at the DHCP server to see both the valid address pools that the DHCP servers will use based on the subnet from which the request came as well as the actual IP address that has been assigned. DHCP servers are then temporarily shutdown, demonstrating that IP addresses can be served by either DHCP server.

2.  Storage Demonstrations

  1. Mirroring – In the disk mirroring demonstration we explore the disk configuration of our servers. All servers have mirrored boot drives. Some servers have additional disk drives which can be used to create new mirrored volumes. Individual drives can be removed from the configuration, simulating a drive failure without loss of data. Virtual drives can then be added back to the system allowing mirrored volumes to be recovered.
  2. RAID-5 - In the RAID-5 demonstration we explore the disk configuration of our servers. Many of our servers have multiple disk drives configured into RAID-5 arrays. Some servers have additional disk drives which can be used to create new RAID-5 volumes. Individual drives can be removed from the configuration, simulating a drive failure without loss of data. Virtual drives can then be added back to the system allowing RAID-5 volumes to be recovered.
  3. SQL transaction logging – All SQL servers have separate volumes to house transaction logs. All databases have transaction logging enabled.

3.  Network Application Demonstrations

  1. Active Directory users and groups – In this demonstration we look at Active Directory Users and Computers.
  2. Active Directory logon scripts via group policy – In this demonstration we look at the login script which is executed by users when they login to the domain. This script maps various drives to file servers based on group membership.
  3. File and Print serving – In this demonstration we look at file and print sharing in an AD network.

4.  Systems Administration

  1. Active Directory password controls – In this demonstration we look at how active directory implements password controls. We can change these controls and examine the impact on the network.
  2. Active Directory auditing controls – In this demonstration we look at how active directory implements auditing controls. We can change these controls and examine the impact on the network.
  3. Patch management with WSUS & group policy – One server in our network is running Microsoft’s Windows Server Update Server, an application that downloads updates from Microsoft on a regular basis and then pushes out updates that have been approved by the systems administrator to workstations and servers that need these patches. In this demonstration we explore the WSUS application.
  4. Network management with ServersAlive – Our network is monitored using a package known as ServersAlive. In this demonstration various network components are shutdown and we see how these events are identified by ServersAlive.