Telecom OnLine: Lecture Mail

 


Welcome to Lesson 12, Day 2- Chapter 11 - Enterprise and Distributed Networks

Course Notes

Lesson 12 Notes - Cont'd

Remote Access Networking

The Remote Access Service (RAS) is included with Windows NT and the Routing and Remote Access Service (RRAS) is included with Windows 2000. Computers using RAS to connect to the network act as if they were directly attached, albeit significantly slower.

T here are two client connection programs, the RAS client and Dial-Up Networking (DUN) software. The connection software used depends on the operating system.

Serial Line Internet Protocol (SLIP)

SLIP is the older of the two connection protocols and was intended for Internet connections. Because of this, if only supports TCP/IP.

Point-to Point Protocol (PPP)

PPP provides a more dynamic mechanism for connecting remote clients. It supports TCP/IP, NWLink, and NetBEUI. It also supports dynamic IP address assignment, which SLIP does not. Also note that on Windows 98, Windows Millennium, Windows NT, and Windows 2000 computers, RAS and RRAS require PPP.

Creating Larger Networks
Repeaters
A repeater is used to regenerate a signal. However, also note that a repeater does not work above the Physical layer so they have no knowledge of the protocols and their structure. If a packet is damaged, but the signal is good, the repeater will regenerate the bad packet.

Repeaters cannot be used to connect different protocols. An Ethernet segment must be connected to an Ethernet segment. However, repeaters are able to connect different cable types, such as Thinnet to UTP.
It is because of propagation delay that the 5-4-3 rule exists for coaxial networks.

Bridges

In addition to connecting network segments and dissimilar physical media, bridges filter traffic and can connect different network architectures.
Bridges also perform filtering. Bridges operate at the Data Link layer of the OSI model. Recall that the MAC sublayer of the Data Link layer is where physical addressing takes place. Using the information contained in packets it processes, a bridge learns which devices reside on each cable segment. It then forwards packets it receives according to the physical destination address of the packet. This process is used by transparent or learning bridges.

Source-routing bridges: These bridges use information provided by the packet’s source to determine the path the packet takes. In a source routing network, the stations do the majority of the work, whereas in a transparent bridging network, the bridge handles determines the packet’s path.

Bridges do not filter broadcast packets. This information is queried on the Microsoft test and it is important the students understand that bridges will not limit broadcasts and will not prevent broadcast storms.

Translation bridges provide connectivity between different network architectures.

Routers

Routers are used to create internetworks, or collections of networks. These internetworks function independently, but are able to share information.

Routers are able to provide fault-tolerance in networks by choosing between multiple paths. If a network is designed with more than one path from Office A to Office B, the router will choose the best path for the packets. However, if the best path is no longer available, the router is able to select another path.

Routers can connect multiple network segments and filter traffic like bridges. It is important to mention that, unlike bridges, routers do not forward broadcasts and can be used to prevent broadcast storms.

Routers operate at the Network layer of the OSI model. Because of this, not all protocols can be routed, only those with Network information.

Because routers operate at the Network layer, they can easily perform the function of translation bridges, i.e., sending information over different network architectures. This is very often the primary task for a router. A good example of this is illustrated in an internetwork which consists of two separate buildings on a campus. The buildings use Ethernet and are connected using FDDI. In this configuration, the router is used to send data between the two buildings and converts all data from Ethernet to FDDI and back to Ethernet.

Routers handle unknown destinations and broadcasts differently than bridges. When a router receives a packet whose destination it does not know, it discards the packet. Routers also do not forward broadcasts or corrupted packets. This is very useful in limiting the effect of network failures and broadcast storms.

Routing tables are used by routers to determine the path a packet must take.The methods of populating these tables include static routing and dynamic routing.

Route discover algorithms: distance-vector and link-state.
Consider a network between three sites. Site A is connected to Site B via a T-1 line, Site B is connected to Site C via a T-1 line, and Site A is connected to Site C via a 56 Kbps line. When a distance-vector algorithm is used to determine the best path for a packet to take, traffic from Site A destined for Site C will be sent directly over the 56K line. However, because link-state algorithms take into consideration network speed as well as the number of hops, when OSPF is used to determine the best path, it will send the traffic through the router at Site B because two T-1 lines are faster than a 56K line.

The spanning tree algorithm is used to eliminate redundant routes. For more information on this topic, see Interconnections: Bridges and Routers by Radia Perlman. Ms. Perlman wrote the spanning tree algorithm and explains the topic in detail and very well. This book is highly recommended.

Routable and non-routable protocols: It is very important to understand that NetBEUI is not a routable protocol. Because of this, it is only suited for small networks using bridges. If a router is installed in a network using NetBEUI, traffic will not be passed between the network segments.

Brouters

Almost all hardware marketed as a router is, in fact, a brouter. Brouters, when correctly configured, are able to perform the functions of both routers and bridges. These devices are very useful in networks using a mix of routable and non-routable protocols, especially those using NetBEUI.

Gateways

These are intricate pieces of equipment which operate at all layers of the OSI model to convert data from one format to another. They are most often found when connecting PC networks to mainframe computers.

Because of the detail involved in making these types of conversions, gateways are slower and more expensive than other networking equipment.

Switches

Switches are being marketed as a new technology. In fact, a switch is a high-speed, multiport bridge. Switches determine path according to the destination hardware address of the packet. Advanced switches that support LAN emulation (LANE), can be configured to route traffic between individual subnets much like routers.

The big difference between bridges and switches is that switches dedicate bandwidth to each connection. For example, in a typical repeater or bridge, all ports in the device share 10Mbps. However, with a switch, each port has what is known as “dedicated 10Mbps.”


Remember to keep up with your on-line lessons and quizzes!

mdm