Network Devices

Devices help control and extend the usable size of a network. Depending on the device, it may join existing networks together,segment or localize traffic on the network,allow a greater number of computers to exist on the network and extend the useable distance of the network.

1. Repeaters:
   • Repeaters are Physical layer (layer 1) devices on the OSI reference model.
   • Repeaters work against attenuation (degradation of signal clarity) by repeating and cleaning up digital signals.
   • Some repeaters can connect two physically different types of cabling but repeaters cannot reformat, resize, or manipulate the data packet.
   • The number of nodes and length of the cable tend to degrad the quality of the signal on a network.
2. Hubs:
   • A hub is a connection device used to tie several networking cables together.
   • Hubs that are plugged into electric power amplify or repeat signals and are called active hubs(aka multiport repeaters). A passive hub and is not a repeater.
   • An intelligent or switching hub segments or subdivides network traffic.
   • A star topology has a hub at its center.
   • Repeaters and hubs increase the usable distance of a network and can connect networks using different physical media.
3. Disadvantages of Repeaters & Hubs:
   • Repeaters do not segment the network.
   • Too many repeaters on a network create noise on the wire and increase the likelihood of packet collisions.
   • Repeaters cannot connect different network architectures.
   • The number of repeaters must be limited.
   • Repeaters do not reduce network traffic.
4. Bridges:
   • Bridges operate at the Data Link layer of the OSI reference model.
   • Bridges do some network segmentation by examining the MAC address that is sent in the data packet.
   • Bridges build tables so that they know which MAC addresses are on the local segments.
   • The bridge reads the destination MAC address of the packet and either forwards or discards the packet based on the information in the table.
   • Bridges reduce collisions by keeping excess traffic off of segments for which it is not intended.
   • Bridges function like repeaters in that they repeat signals but bridges selectively repeat a packet based on the MAC address.
   • Broadcast packets are forwarded by the bridge so bridges define collision domains but not broadcast domains.
5. Bridge types:
   • Transparent Bridging: Transparent bridges are also called learning bridges because they build their table of MAC addresses as they receive packets. When a bridge first receives power, its bridging table is empty, but over time, it learns which segments have which MAC addresses as packets are forwarded. The bridge uses the source address to determine which addresses are on which segments. Ethernet networks mainly use transparent bridges.
   • Source-Routing Bridges: Source-routing bridges rely on the source of the packet transmission to provide the bridge table information. The source computer determines the best path by sending out explorer packets. This information is used by the bridge to build its table. The destination determines the best route for future packets to take. The destination returns that information to the source. Token Ring networks mainly use source-routing bridges.
   • Translation Bridges: Bridges usually connect network segments with similar architectures. A translation bridge can connect segments with different architectures such as Token Ring and Ethernet.
   • Advantages of Bridges: Bridges can extend the network by acting as a repeater. Bridges can reduce network traffic by subdividing network communications. Bridges reduce the likelihood of network collisions. Some bridges connect networks using different architectures.
   • Disadvantages of Bridges: Since bridges do more than repeaters by viewing MAC addresses, the extra processing makes them slower than repeaters. Bridges forward broadcast packets indiscriminately, so they do not filter broadcast traffic. Bridges are more expensive than repeaters. The fact that bridges forward broadcast traffic is a major disadvantage. Since bridges simply forward broadcast traffic, they contribute to broadcast storms.
6. Switching Hubs:
   • Switching hubs operate at the Data Link layer of the OSI reference model. They increase network performance by reducing the number of packets transmitted to the rest of the network. The switch opens a virtual circuit between the source and the destination computer. This prevents communications between two computers from being sent to every computer on the network or segment. When two machines have a virtual circuit between them, they do not have to share the wire with any other computers.
   • Advantages:1) Decreased packet collisions 2) Increased available network bandwidth 3) Increased network performance
   • Disadvantages: 1) Broadcast traffic may be troublesome on a switch. 2) They are significantly more expensive than bridges. 3) Network connectivity problems can be difficult to trace through a switch.
7. Routers:
   • Routers operate at the Network layer of the OSI model.
   • Routers have the ability to connect multiple segments and multiple networks.
   • Networks connected by multiple routers are called internetworks.
   • Routers use the logical address of the OSI Network layer.
   • Routers do not forward broadcast traffic and therefore define broadcast domains.
   • Routers maintain routing tables in RAM to keep track of network IP addresses.
   • When a network uses TCP/IP, each port (aka interface) on a router requires an IP address.
   • Advantages of Routers: 1) Routers can reduce network traffic because they do not retransmit network broadcast traffic. 2) Routers can connect different network architectures, such as Ethernet and Token Ring. 3) Routers can choose the best path across the network using dynamic routing techniques. 4) Switches are the device of choice for improving performance on a network, routers are used to segment large networks into internetworks made up of smaller networks.
   • Disadvantages of Routers: 1) Routers are more expensive than bridges or repeaters. 2) Routers work only with routable network protocols 3) Routers are slower than bridges or repeaters because they must analyze a data transmission from the Physical to the Network layer. 4) Dynamic router communications causes additional network traffic.
8. Physical vs. Logical Addresses:
   • The NIC manufacturer burns the physical address or MAC address into the network card during the manufacturing process.
   • The MAC address is usually represented by 12 hexadecimal numbers.
   • TCP/IP and IPX/SPX protocols are routable, meaning that they can be used for communication between network segments divided by a router or routers.
   • Routers utilize the IP address to route packets to the correct network segment.
   • The MAC address is implemented in hardware but the logical address is implemented in software.
   • Bridges and switches use the physical address at the Data Link layer but routers use the logical address at the Network layer.
9. Dynamic & Static Tables:
   • Routing tables can be built statically or dynamically.
   • A static routing table requires manual configuration by the network administrator and the router always uses the same route to send packets to a specific network address, even if that route is not the most efficient at the time.
   • Routers automatically build dynamic routing tables.
   • The router communicates with other dynamic routers to determine the most efficient route from one network to another.
   • Dynamic routers can use distance-vector algorithm which takes into consideration the number of hops between two points or link-state algorithm which takes into consideration network traffic, connection speed, and other assigned costs.