NETWORK CONCEPTS

Advance Hardware :
1. What is attenuation? Attenuation refers to the weakening of the signal as it travels over cable. It is sometimes referred to as roll off. As a signal travels across wire, its square wave becomes deformed in proportion to the distance traveled. Thus, attenuation is primarily a function of cable length. If the signal travels too far, it can degrade so much that the receiving station may not be able to interpret it and communication will fail. A repeater is can be used to boost signal strength on a baseband network so that it can travel greater distances. An amplifier performs the same function on a broadband network. 2. What is a repeater? A repeater is a device that regenerates the signal transmitted on a cable. Repeaters allow signals to travel beyond the normal cable length limitations. A repeater does not translate or filter packets. Repeaters have the following characteristics: sed to regenerate an e!isting baseband signal sed primarily in a coa!ial bus "linear# topology $egments connected by a repeater must use the same media access control "%A&# method "e.g., a repeater does can not pass traffic between 'thernet and Token Ring# A repeater can pass traffic between different types of media "i.e., coa! to fiber optic# if the appropriate interfaces e!ist $egments connected by a repeater must have the same network address A repeater does not accelerate or change the signal( it simply regenerates it A repeater does not filter packets or limit congestion A repeater will pass a broadcast

A repeater operates at the physical layer of the )$I model The basic functionality of a repeater "signal regeneration# can be incorporated into other connectivity devices "e.g., a hub or bridge#

*ecause a repeater does nothing to filter or limit traffic on the network, it should be viewed primarily as a means of connecting distant workstations, not as a means of adding additional workstations. In other words, repeaters should be used to e!tend the length of a network but not its density. 3. Is there a limit on the number of repeaters that can be employed? +es. The number of repeaters is limited by the ,-./.0 rule., nder this rule there can be no more than - segments with no more than / repeaters between any two stations. )nly 0 of the - segments can be populated "i.e., contain nodes or hosts#. In this case a ,station, is a bridge, router, or gateway. Thus, adding a bridge can allow additional repeaters to be used on the 1A2. What is a hub? A hub is a generic term used to describe a central connection point for computers on a network. The basic function performed by a hub is to receive signals from one computer and transmit them on to other computers. A hub can be active or passive. An active hub acts as a repeater( it regenerates and sends out a stronger signal. A passive hub is basically 3ust a 3unction bo!( it splits incoming signals for transmission across the network. The hub is the center of a star topology and allow computers to be added or moved on the network with relatively ease. How does a hub work? A hub is basically a signal splitter( it takes the incoming bits from one port and sends them out to all of the other ports. A eight port hub would receive a packet on one of its ports and send copies out to each of the other seven ports. 'ach host connect to the hub will view the packet but only the host to which it is addressed will process it. This can cause network traffic problems because a packet addressed to one host is actually sent to all hosts "even though it is only processed by the one it is addressed to#. What is switch? A switch is an 4intelligent hub4 that has the ability to determine the destination %A& address of a packet. Instead of passing the packet to all ports, the switch forwards it only to the port to which it is addressed. Thus, a switch can drastically reduce network traffic A switch maintains a list of %A& address associated with its various ports that it uses to determine where to send the packet. *ecause it operates on the %A& address rather than an I5 address, a switch is generally faster than a router. 8. What is a M !"?

A %$A is a %ultistation Access nit that functions like a hub in a Token Ring network. A %$A has the ability to bypass a cable break or station failure in order to maintain the integrity of the ring. What is a concentrator? A concentrator or wiring concentrator is similar to a hub and the two terms are often used interchangeably. It serves a central connection point where a number of incoming cables converge. A hub or switch performs the function of a concentrator. In its most simple form, a concentrator is a multi.port repeater. It merges a number of incoming lines with a number of outgoing lines. $ome concentrators can control errors, determine the quality of cable, and perform other administrative functions. 1ike a %$A the concentrator in a ring network keeps the ring intact even if workstations are removed. What are #interconnecti$ity# de$ices and what functions do they perform? A concentrator or wiring concentrator is similar to a hub and the two terms are often used interchangeably. It serves a central connection point where a number of incoming cables converge. A hub or switch performs the function of a concentrator. *ridges, routers, and gateways are interconnectivity devices. Their primary function is to connect segments or multiple networks into a single, heterogeneous system across which computers can communicate. These devices are sometimes referred to collectively as ,relays., What is a brid%e and what can it do? A bridge is a relay or interconnecting device that is use to 3oin to 1A2s. *ridges, routers, and gateways are interconnectivity devices. Their primary function is to connect segments or multiple networks into a single, heterogeneous system across which computers can communicate. These devices are sometimes referred to collectively as ,relays., A bridge can: 1ink segments of a network together 6ilter traffic by reading packet addresses 1ink dissimilar networks "'thernet and Token Ring# if higher level protocols are the same or can be translated

A bridge cannot: 7etermine the most efficient path to transmit data 5rovide traffic management functions "congestion bypassing# The functions of a bridge may be performed by a server or an independent bridge device. A general rule of thumb is that there should be no more than four bridges on a 1A2.

How does a brid%e pass traffic between &!' se%ments? A bridge checks the %A& address of each packet it receives. The bridge e!amines the packet4s destination address and compares it to the routing table it maintains. *ased on this comparison, the bridge performs one of three tasks: If the destination address is in the routing table and is located on the same segment as the sender, the packet is dropped. "There is no need to pass it across the bridge if it is addressed to a host on the near side.# If the destination address is in the routing table and is located on a different segment, the packet is forwarded to that segment. If the destination address is not in the routing table, the packet is forwarded to all segments.

2ote that when a bridge receives a packet from one host that is addressed to a host on the other side, the bridges passes the data frame across the connection. 8hen the bridge detects traffic that is addressed to a host on the originator9s segment, it does not allow it to pass. In this manner a bridge performs a filtering function that reduces the overall traffic on the network. Although a bridge can learn the %A& addresses of the stations on the network, it cannot determine the most efficient data path to send the data. This 3ob requires a router. How does a brid%e create a routin% table? The routing table for a bridge is different than a routing table used by a router. The routing table used by a bridge is based on %A& addresses rather than I5 Addresses. nlike routers, bridges don4t 4talk4 to each other to update their routing tables. :owever, most bridges can monitor and ,learn, the addresses of each station on the network. They use this information to construct a routing table. 8hen the bridge receives a packet, it e!ams the packet4s source %A& address and uses this information to add to or update its routing table. A bridge only needs to know which %A& addresses are on which segments so it can properly forward the packets. 8hen it e!amines a packet, it may not know on which segment the destination address resides, but it is does know from which segment the packet originated. $o the source %A& address and segment are added to the routing table. )ver time the routing table will be updated with all %A& addresses and segments. What is network partitionin%? If network performance is being degraded by e!cessive traffic, one possible solution is to partition the network into two or more segments. A network can be partitioned by using a bridge. The bridge is inserted between two segments of the 1A2 and as traffic flows across the 1A2, the bridge filters and forwards it according to the address. )nly traffic destined for another segment is allowed to ,cross, the bridge. 5artitioning is also referred to as segmenting a network. How does partitionin% (se%mentin%) benefit a network?

8hen e!cess traffic starts to degrade performance, one solution is to partition or segment the network. 6or e!ample, assume that a company network connects the accounting and marketing departments which generate the following number of messages during the course of a typical day: %arketing to %arketing: ;<<( %arketing to Accounting: =<<( Accounting to %arketing =<<( and Accounting to Accounting: ;<<. 8ithout partitioning, all traffic will flow across both departments. Thus, ><< messages will be directed to each host. :owever, if the two departments are partitioned by separating them with a a bridge, all of marketing9s interdepartmental messages will stay on their side of the network. )nly messages addressed to accounting will cross the bridge to the accounting segment. The reverse is also true( the marketing segment will not be cluttered by internal messages from accounting. The result is a -<? decrease in overall network traffic( each segment now only processes 0<< messages. As a general rule, if all you want to do is partition a network, a bridge is a better choice than a router because it operates with lower overhead. What are some of the potential problems with connectin% *thernet and +oken ,in% &!'s with a brid%e. *ridges are capable of connecting 1A2s that use different physical and %A&.layer protocols, such as 'thernet and Token Ring. :owever, there are several potential problems with these connections: Token ring has a mechanism for setting transmission priority. 'thernet does not.@ 'thernet packets are smaller and structured differently than Token Ring packets 'thernet packet information is encoded differently than Token Ring packets 'thernet uses transparent bridging to identify network addresses and token ring uses source routing

@=<<AB.Any1A2 'thernet can establish transmission priorities but by an different method than that used by a Token Ring network. A translational bridge can be used to eliminate may of the problems associate with linking 'thernet and Token Ring 1A2s "e.g., packet format differences, address detection methods, etc.# Among other things, this type of bridge is able to remove data from one frame type "e.g., 'thernet# and repackage it into another frame "e.g., Token Ring#. What are the benefits pro$ided by a brid%e? A bridge is a relay or interconnecting device that can be used to provide the following capabilities: '!pand the length of an e!isting network Increase the number of workstations on the network Reduce traffic congestion "network partitioning#

5rovide a connection to a dissimilar network "e.g., 'thernet to Token Ring# %ove data across a intermediate network with a dissimilar protocol

What is transparent brid%in% and how does it compare to source routin%? +ransparent brid%in% and source routin% are two learning techniques used by bridges to build a table of addresses for all of the stations on a network. 8ithout these techniques, system administrators would have to load the addresses into the bridge manually. +ransparent -rid%in% is used in 'thernet 2etworks. *ridges learn station addresses by e!amining data frames as they pass. The address table associates the source address of the data frame with a network address. The table is reviewed continually and updated as the network changes. ource ,outin% is used in Token Ring networks. *ridges learn station addresses by having a source node send out e!plorer packets. %ultiple e!plorer packets are sent out over the network to their destination. They then report back to the source node which determines the most efficient path. The path information is stored in the bridge and all subsequent transmission use this path.

Ceep in mind that the ob3ective of both transparent bridging and source routing is to prevent bridges from creating loops in a network. The distinction is that transparent bridging is an 'thernet method that blocks loops and source routing is a Token Ring method that avoids loops.

Home Forum-Help Mail Box & Other Services Bookmark (Ctrl + D) Co tact !s Sitemap " #$$% - - IT Acumens & Free 'esource Shari ( )lace For * (i eers+