Sandbox‎ > ‎Archive‎ > ‎IPT 2008-09‎ > ‎Blair's page‎ > ‎Blair's updates!‎ > ‎

2009-03-17 Networking technology

posted Mar 16, 2009, 12:56 AM by Unknown user   [ updated Jun 12, 2009, 6:40 PM by Eddie Woo ]

Class notes - 'Network devices'

This is a transcript of handwritten notes dated 2009-03-17.

'What kinds of hardware are likely to inhabit a network?'

 Network Interface Card (NIC)
 Like an airport (network interface port)
 Controls sending, receiving, and error check on the client computer
 Cables, backbone
 Cables are reliable, fast, and inexpensive
 Backbone is used for situations with more traffic
 Repeaters Repeaters are like 'megaphones', allowing for an input to be increased
 in signal strength
 Hubs, switches, and routers
 Any of the three can be the central component of a star network
 Gateways Analogous with interpreters or embassies
 e.g. in a hybrid-topology network, a gateway may allow a node on a ring
 network to interact with a node on a star network
 Wireless-specific WLAN stands for Wireless Local Area Network
 There are access points or 'hotspots' on a WLAN
 Celluar networks are an example of wireless networking (but not local)
 Nodes need wireless adapters (NIC)

Compare and contrast hubs, switches, and routers

Morning class: Compare and contrast hubs, switches, and routers. Why would you choose one over another?


When a node sends data to a hub (Ethernet hub), the hub will send it to all the other clients on the network, regardless of which clients the data is actually addressed to. Clients to which the data is not addressed simply ignore the data when it arrives. This works in a similar way to a public announcement system, and similarly, is useful when the majority of data that is typically sent in the network is usually addressed to most or all of the nodes.

(Admittedly, in the above illustration, the hub (not shown) in the second scenario could only be replaced by a network switch if the node that belongs to the owner of MDKP-56 is identifiable on the network.)


Because there are many network setups in which the data is directed at individual nodes and not to groups of nodes, switches are designed as an alternative to hubs. They analyse the data to determine which node is the recipient, and then forwards the data to that node only.


Routers can be thought of as more thoughtful switches. They are suited to more complex networks where not all nodes are connected in the same way to the router. For example, the network that a particularly router works on may have a node not directly connected to the router. When the router receives data intended for this node, it will dynamically determine the quickest path currently available to the node, and then channel the data through this path instead of another path, which may, for example, be experience high amounts of traffic at the time.

Wireless networking advantages

Morning class: Discuss the advantages of WLAN technologies in the medical, commercial, education, and engineering solutions.


A nurse may be able to physically be helping one patient, but requested at any time by someone with a more urgent case, if the nurse is equipped with a communication device with wireless networking. This would eliminate the need for a public announcement system. Both doctors and nurses would be able to access databases containing all the patient information they would possibly need with a physically small device, eliminating the need for often impractically large bundles of documents. Furthermore, as medical tools become increasingly more reliant on information technology, wireless networking could, in fact, be applied in medical tools themselves; for example, nanobots that may one day be invented to operate on a microscopic level would need to be controlled wirelessly, as a wired connection would limit the versatility of the nanobot.


Already, couriers are being equipped with small communication devices, requesting signatures on delivery not by pen and paper, but by stylus and touchscreen. They are able to immediately send a notice of confirmation to the sender, through a wireless internet connection.


The state government of NSW appears to be interested in the concept of installing wireless network hotspots in schools themselves, which would allow for portable network access. This has the benefit of allowing information technology to be used in places where it would otherwise not be practical or sensible, and would also eliminate the need for expensive, high-security computer "labs".


Some engineering projects are now very physically big. For example, the Large Hadron Collider, an apocalyptic physics experiment apparatus in Europe, is 27 kilometres long, 9 kilometres wide, and up to 175 metres deep. In order to build and operate this machine, individuals that were stationed around the area would have needed to communicate with each other. In this situation, WLAN would be preferable over a wired solution as the individuals would be free to move about while still maintaining contact, which would be advantageous in a teamwork situation requiring simulatenous collaboration and physical work.

Network services

For further details, please refer to IPT: Network Services
  1. A network service is software on a server that automates the process by which the server gives clients access to shared network resources, such as printers and shared files.
  2. Examples of network services are printing and file sharing.
  3. Network services give clients access to shared resources - often, there are multiple clients. In order for the network service to know which client is requesting access to which shared resource, clients must be identifiable. In order for clients to be identifiable, an address service must exist.
  4. DHCP is an acronym of Dynamic Host Configuration Protocol. As implied by its full name, DHCP is a protocol through which the server automatically and actively gives clients instructions as to how to configure themselves to operate on the same network as the server. It generally does this by automatically allocating all new devices a unique IP address, 
  5. Advantages include less data storage space consumption, less initial setup time (no need to install software on a large number of computers assuming that the network itself is large), and ease of upgrading/reinstalling/repairing software setups; disadvantages include slow access speed (software would need to be executed via a network), inconsistency (server workload changes depending on how many clients there are and what the clients are doing), and potential unreliability (clients would become dependent on the server, which may or may not be prone to downtime)
  6. Records: must contain every single person who could possibly have access to the network. Fields: date of entry into the system, access level, possibly data to identify who the person is, depending on the kind of network, possibly computer ID/IP address, depending on whether computers are shared. Attributes: date of entry to the list of changes to the network, computer ID from list of computers, IP address from network configuration database.
  7. If a network did not have user accounts, it would be nearly impossible to manage the varying levels of access to network resources. This particularly applies to file sharing - read/write access to files stored on a network are frequently restricted, usually in the name of privacy and security.
Eddie Woo,
Mar 17, 2009, 3:25 AM