The material presented in this module is basically what I like the call "Pre-course material" or basically the stuff you are supposed to know

In order to survive, you should study and attempt to memorize each bullet point, there is valuable information contained in each one that will carry over to the next module, or modules, in some form or fashion.

• What is a Network and why do we need it? - The answer is easy, but lets start with why we need it. Imagine you had to print something out, wouldn't it be easier to just press the print button and then walk to the printer and Blam!!! there it is? The alternative without a network would be to go to another computer with a zip or thumb drive, possibly a floppy disk or CD and use someone elses computer to print it out. Now you have two people tied up over one print job. From the management perspective, buying a printer for everyone isn't economically feasible either.  So the reason we need networks is to share resources and information and a network is two or more computers connected together sharing a resource. Common items found on a simple network are computers, printers, switches, data storage, information, etc...
• What is an Internetwork? - and internetwork is two or more networks connected together and what connects networks together to form and Internetwork...a router of course!!!
• What is some of the equipment that we use in networking? -
  
  • NIC - network interface card - installed inside the computer with protocols that are " bound" to the NIC. Current standards are ODI (Novell) and NDIS  (Microsoft).  NICs are found at the 2nd layer of the OSI model and have a MAC Address burned into the card.
  • Cable
  • Terminators
  • T-Connectors
  • Barrel Connectors
  • Repeaters - one of the first attempts at a hub and is used to forward or regenerate a signal from one segment to another. Similar to a bull horn making a persons voice louder. Just like a bullhorn, repeaters will also regenerate or increase any line noise that it may have picked up on the way. Repeaters are found at the 1st layer of the OSI model.
  • Hubs - commonly referred to as multi-port repeaters as they do the same thing as a repeater but have more ports on them. Hubs are still in use today and highly NOT recommended by this author. You'll learn why later. Hubs will forward any and all signals it receives out all ports except the one it came in on. Hubs also use shared bandwidth, which means that if there are 10 computers on the network and all 10 are transmitting then only 10% of the performance will be seen...assuming a broadcast storm doesn't start up. Hubs provide for an easy solution to add nodes to the network without having to run cable and are found at the 1st layer of the OSI Model. Hubs are used to create a physical star network while maintaining a logical linear bus network similar to 10base2.
  • Bridges - Similar to a repeater, but more intelligent by having a bridging table. A bridging table allows for the bridge to learn the network and what devices are connected to it. As data comes into the bridge, the bridging table is populated with the MAC addresses of the source computer and what port the data came in on. As future data comes in the bridge knows exactly what port to forward the traffic out of in order to reach the destination computer. It is important to understand that this technology is done by MAC address only and not IP addresses. Because of this a bridge is not as intelligent as a router, yet performs some of the same functions as a router just at a lower level. Bridges are found at the 2nd layer of the OSI Model. Common problems with bridges are bridging loops when two or more bridges are plugged into the same networks. Bridges will help limit the size of the collision domain but are not considered the prime piece of equipment to do so.
  • Switches- similar to a repeater and a bridge in that a switch will forward data out all ports like a repeater but also learns about the network the longer it is turned on like a bridge. Similar to a bridge, a switch will learn the source MAC address of the sending computer as the data comes into the switch and can send the data out the correct port to the destination computer rather than all ports. Switches don't reduce traffic, but they do optimize traffic by reducing the size of the collision domain to just two devices (source computer and destination computer). Switches are found at the 2nd layer of the OSI model and some switches are capable of layer 3 functions (routing) to make them multi-port routers. Switches are also one of the three required components for Full Duplex data transfer (switch, full duplex capable NIC, and collision detection turned off).
  • Routers - Routers are found at the 3rd layer of the OSI model and are used for a variety of reasons :
    • Routers segment the network and make it more managable and controllable. If a network had 200 users on it and broadcasting was getting out of hand then a router could be used to split the network in half to correct the broadcasting issue.
    • Routers control broadcasting by a variety of methods.
    • Routers are used to connect LANs to form MANs or WANs.
    • Routers limit broadcasting domains.
  • Gateways - Gateways are used to connect different types of networks together such as a PC based network to a mainframe. Gateways are first found at the 3rd layer of the OSI model and found in higher layers as well. All email servers are considered to be a type of gateway considering that the source and destination computer (and email server) are unknown for virtually every email that gets sent.
  • Servers - Servers are the cornerstone of where the data is held, how the data is stored, how the data is manipulated in relation to printing, etc... Servers perform a variety of functions also and can be specialized, for example, you can have email servers, file servers, print servers, backup servers, Anti-Virus Servers, media servers, the list will probably never end in what functions servers can do. Some servers are even configured to be routers as well although that is HIGHLY not recommended.
  • Firewalls - Firewalls are used to connect networks of any type ( LAN, WAN, MAN, GAN, etc...) to the Internet. Firewalls can also be used to compartmentalize departments within a network to protect sensitive information from employees. Firewalls perform basic functions like NAT, PAT, VPN, etc...

• What type of equipment is used in a WAN?
  • Routers
  • ATM Switches - ATM switches work just like the previously mentioned switches "except" ATM networks don't have packets they have cells. Because of this ATM switches are faster than normal switches because each cell is 53 bytes in length. How does a packet go from normal style packets to an ATM cell? Through a router of course!!! ATM switches are normally found only in large organizations or telephone companies.
  • Frame Relay Switches - Same concept as an ATM switch but running Frame Relay instead. Frame Relay will be covered in future modules.
  • X.25 Switches - Same thing as a Frame Relay switch but running X.25. Not really found any more as most have upgraded to Frame Relay.
  • Modem - Thats right, modems are still in use today. Modems convert digital signals into analog signals so they can be sent over a phone line.
  • CSU/DSU's - similar to a modem as an adapter but doesn't do any conversion. CSU/DSUs are used to terminate a digital circuit such as a T1 line at the customers site. CSU/DSU used to be the size of an external modem (some still are), but most have been shrunk where they are in card format and installed inside the router.
  • MUX/DeMUX - used to combine multiple channels into one physical channel and vice versa at the other end.
  
  
• What kind of media/wire is used in a network? There are so many different ways of connecting devices it's insane. In relation to cable you have to be aware of shielded or unshielded, plenum rated or not, the pin outs (straight through, roll over, or cross over), etc... The various types of media also have their issues such as price, attenuation, noise or EMI, crosstalk, distortion.
  • Coaxial - both type of coaxial cable require 50ohm terminators at the ends of the cable in order to terminate the signal. Collisions will result if there is a cable break or a terminator missing.
    • thinnet - commonly called 10base2
    • thicknet - commonly called 10base5 but can also be called "ugly yellow garden hose" as the cable is normally an ugly yellow cable about the size of a garden hose.
  • Twisted pair
    • CAT3 - only seen for 10baseT networks and only allows for a maximum speed of 10MBPS
    • CAT5 - CAT5 comes in several varieties and the most common one seen today is CAT5e (e meaning "enhanced") to allow for higher speeds. Most CAT5 networks are 100MBPS.
  • Straight through - a straight through cable is used to connect a PC to a hub or a switch and is the most common cable in use. Some of the more popular lower end home networking devices are auto sensing in relation to straight through or cross over. Cisco is not though and for the purposes of the test or any labs you must be aware of the type of cable you are using. If you were to take a straight through cable and compare both end, they should look exactly the same.
    
  
  
  • Cross over - a cross over cable is used to connect two like devices together such as a switch to a switch or a router to a router or a computer to a computer. This is done to allow the transmit and receive pins to be matched up correctly. On a cross over cable pins 1 and 3 are switched as well as pins 2 and 6.
  • roll over - a roll over cable is primarily only used to connect to a routers console port. Should you compare both ends of the cable then they should look like mirror images of each other.
    
  • Fiber Optic - fiber is composed of either glass or plastic and is the most expensive, fragile, and secure form of connecting devices together. With fiber optics data travels down the cable and "tapping" the line is extremely difficult if not impossible. Most of the issues that wire has are not found with fiber with the biggest benefit being EMI free...as in, EMI is not found in fiber. Light and electricity simply do not mix.
  • RF - Radio Frequency is becoming more and more common nowadays with the sudden influx with wireless devices and also on the CCNA test. However, wireless does have it's flaws :
    • broadcasts in all directions and allows outsiders to sometimes gain access. Gaining access is commonly called "War Driving", where someone with a laptop drives around until they find a signal and then surfs the Internet for free or steals data found on your hard drives.
    • Generally not secure because most wireless devices don't have the encryption turned on by default. Back to the war driving scenario again....
    • Wireless can be more secure by using frequency hopping but freq. hop. is generally only found in the pricier equipment.
    • RF can sometimes require an FCC license to use.
    • RF does allow for a high degree of mobility but only within the range of the signal.
  • Laser - Now we are getting a little bit into the realm of James Bond but laser connectivity can be found. Laser allows for a very secure method of communicating without having to have a T1 line or fiber optic line...unfortunately they are also very expensive!!!
  • IR - Infrared is becoming more and more popular with data speeds in the GB range. IR is essentially the same thing as laser but in a cone fashion rather than a little red dot. IR has limited distance and slightly less secure but very effective.
  • Microwave - Same as IR but using microwaves. One of the benefits of Microwave is that it hardly goes down...not sleet, snow, rain, etc...will keep us from our appointed rounds. I have seen microwave dishes continue to function with no problems through hurricanes!!!

• What kind of physical topologies can we find in networking? Well there are a LOT, but most of you will only see the physical star topology. Considering though that you might run across another type, we'll go over all of them here. You have to understand though that we are talking about PHYSICAL topologies...not logical topologies
  
  • Bus - the bus network was the first real network that we need to know about. No, for you old people I'm not going to talk about ARCNET or Banyan Vines. Go to a museum if you want to see that stuff. For our purposes lets associate the bus network with thinnet (aka 10base2)considering that's where most of you will be seeing it. Thinnet had several characteristics that formed the cornerstone of networks and topologies and are listed below.
    • uses CSMACD for media access
    • IEEE specification for CSMACD is 802.3
    • Follows a linear format meaning that the cable literally starts at one end and goes to the other with computers plugged in on the way.
    • more than 30 computers generally degrade network performance to WAY below an acceptable level.
    • Collision is the most common error to occur and then broadcasting
    • Cable break or missing/malfunctioning terminators will result in massive network failures.
    • require 50ohm terminators at each end.
    • maximum segment length of thinnet is 185 meters
    • maximum network length of thinnet is 925 meters.
    • maximum segment length of thicknet (aka 10base5) is 500 meters
    • maximum network length of thicknet is 2500 meters
    • 
      

    

  • Star - The star network is the most common network in use today. All the devices are plugged into a central device and form a star shaped network with a hub or a switch in the middle. Star networks also have some attributes that you need to know about and they are listed below :
    • Even though it is a STAR network and not a BUS network, this type of network topology still uses CSMACD as a media access method.
    • A benefit over the BUS network is that if there is a cable break then the entire network doesn't crash.
    • 50ohm terminators are not required in order to function.
    • Collisions can still happen but can be ruled out if using a switch instead of a hub.
    • Maximum number of nodes on a network is 1024 but don't ever go that high.
    • Performance hits will start to be seen at around 200 computers rather than 30 as found in a BUS network.
    • Commonly called 10baseT or 100baseT (we'll go over this later.)
    • Maximum distance of any one cable can not be over 100 meters.
  • Ring - Ring networks are cool!!! But, alas, not really seen that much any more. IBM is really the first company to implement them, as well as copyright that technology. What happens when you use copyrighted technology? That's right, you have to pay royalties...which led to the ultimate demise of most ring networks. The attributes are listed below :
    • Uses token passing as a method of transmitting data
    • IEEE standard is 802.5
    • IBM is/was the biggest proponent
    • very reliable technology
    • had self healing characteristics
    • physically looks like a STAR network but logically it is a ring.
    • A common ring network that is still in use is FDDI (Fiber Distributed Data Interface) which uses fiber optics for the ring and is very expensive.
  • Mesh - the mesh network is for the ultra paranoid!!! Mesh is pretty much a network where there is more than one connection to the same place. This allows for redundancy in case of a cable or hardware failure. Redundant backlinks are becoming more and more popular today with companies wanting to expand and branch out to different parts of the world. Most core level devices in a medium or large company will have a mesh network to connect everything together.
  • Hybrid - any combination of the previously mentioned topologies.
• How are networks described? Networks are generally described by their capabilities, as in how fast they go. Common terms are 100baseT or 10base5 or 10baseT. This naming convention is known as SPEED TYPE LENGTH. To start with the beginning, lets discuss 10base5. You should also understand that some technologies aren't so obvious...like 10base2 or 10baseT (how does the letter T equal to distance anyway???).
  
  • 10base5 (aka thicknet) means that the network is 10MBPS, baseband (covered later), and segment length is 500 meters.
  • 10base2 (aka thinnet) means that the network is 10MBPS, baseband, and 200 meters in length. Or is it??? In this case they rounded up, the max length of a 10base2 segment is 185meters. I know it gets confusing sometimes.
  • 10baseT is where the first case of warping the definition comes to play. 10baseT means 10MBPS, baseband, and Twisted pair as a cable. Also the length of that cable can not be longer than 100 meters. 10baseT is the first occurence of taking a logical BUS network and turning it into a physical STAR network.
  • 100baseT same as 10baseT but 100MBPS.
  • 100BaseVG - this is an old technology that allowed for 100MBPS transmission over CAT3 cable. The intention was to allow for companies to upgrade their connectivity without having to replace their existing wiring. To bad that didn't work out!!!
  • 100baseFX - same thing as 100BaseT but using Fiber Optic cable.
• What rules do we have to follow when networking? Well there are quite a lot. W've all ready spoken about maximum segment length depending on what type of topology we are using. Probably the most basic and biggest rule to start off with is the 5/4/3 rule. The 5/4/3 breaks down into segments, hubs/ repeaters, populated segments. Essentially this means that from one computer to another there can only be up to 4 hubs/ repeaters between them, no more than 5 total segments connecting them, and 3 of those segments can be populated. The 5/4/3 rule is OLD!!! But, in some cases, still applies. The 5/4/3 rule can also be broken with the proper application of layer 3 devices such as a router. How else were you able to read this article if it wasn't for a router that allowed you to violate the 5/4/3 rule. Oh, that sounds sooo dirty!!!
• What type of transmissions can we have over our networks? There are quite a few different methods of communicating over networks now. And, with everything moving to IPv6, there will be more!!!
  • Unicast - A unicast is a directed transmission from one device to another. Unless this transmission is filtered out by a router or firewall then it will pass through. An example of a unicast transmission is when you ping a device
  • Multicast - a multicast is a transmission sent to a specific subset of computers. Multicasting uses class D IP addresses and can be found in common applications such as media streaming or using ghost.
  • Broadcast - a broadcast is a transmission sent to all nodes on the network. Layer 2 or below devices will propagate broadcasting however layer 3 devices and higher will prevent broadcasting from passing through. Routers are designed to stop broadcasts. Too many broadcasts will result in a broadcast storm. A broadcast address can look like the following :
    • 255.255.255.255 - Decimal in the form of an IP address (layer 3)
    • ff.ff.ff.ff.ff.ff - Hexadecimal at the MAC address level (layer 2)
    • 1111.1111.1111 - Binary (layer 1)

That is the end of the first module and, as I indicated in the beginning, is intended solely for you to be in a position to understand the rest of the modules with less difficulty. Most students jump into the deep end of the pool and try to come up with some master memorization plan. Understanding the basics is extremely important in order to understand the advanced material. Understand this material first, when you can roughly recite it then you are ready for the next module and to jump right into the OSI model!!!

Please rate this article at the top....if you have any specific suggestions then please email me at This e-mail address is being protected from spambots. You need JavaScript enabled to view it