Ethernet Standards and Cable Types

The Need for and Operation of CSMA/CD

CSMA/CD = Carrier Sense Multiple Access /Collision Detection. We no longer need CSMA/CD so much because each switch port is now it’s own collision domain. However in some older networks where they are using hubs, or where they are using coax for example, CSMA/CD is required. When any host on a shared segment wants to transmit data it listens on the wire for any data – if it detects signals, it backs off for a random period of time before listening again, however if it does not detect anything, it attempts to send the data.

Issues occur when more than one host listens at the same time and detects no signal, so they both attempt to send data. This will cause a collision, and when you have a collision, both data signals are now useless. Normally, when a collision occurs, the voltage on the wire changes because essentially both voltages have hit each other. Now when CSMA/CD is in use, the hosts will detect this voltage spike and generate a JAM signal. So now both hosts that were involved in the collision are going to invoke a random backoff timer, before they attempt to resend the data. Once the backoff timer has expired, both hosts will go through the listen and transmit process again.

Ethernet Types and Standards

10Base-T is specified by 802.3 and the T stands for twisted pair cable. The maximum length of a 10Base-T cable is 100m. Twisting pairs of wires inside the cable cuts down on the possibility of electromagnetic interference.

The first Ethernet standards were 10Base-5 and 10Base-2.

Ethernet is considered a logical bus topology.

Fast Ethernet is defined by 802.3u. Defined by 802.3z and 802.ab is Gigabit Ethernet. Gigabit Ethernet has a maximum capacity of 1000MBPS.

Standard Ethernet cabling type is Category 5 UTP(Unshielded Twisted Pair) and is normally terminated with a RJ-45 connector. In Cat 5, pins 1 and 2 Transmit, and 3 and 6 Receive.

Crosstalk is caused by electromagnetic interference. Essentially,a signal crosses over from one pair of cables to another, causing the signals to become unusable. NEXT is Near End Cross Talk and and FEXT is Far End Cross Talk. PSNEXT is Power Sum Near End Cross Talk.

Ethernet Addressing

Ethernet Addresses are MAC Addresses – MAC stands for Media Access Control. The MAC Address is a 48 Bit Address expressed in Hex. The first half of the MAC is the Organisationally Unique Identifier(OUI), the second half is chosen by the vendor that created the NIC. The MAC address is usually burnt into the card. There are broadcast and multicast MAC Addresses.

The broadcast MAC Address is all F’s. FF-FF-FF-FF-FF-FF. There is a range of Multicast MAC addresses and the first half of a Multicast MAC Address is always 0100.5e. The second half of a multicast MAC address will fall in the range 00-00-00 to 7F-FF-FF.

CCNA Study Notes – Chapter 1

CCNA Study Notes – Chapter 1

Networks

When designing and maintaining a network, remember these factors:

Cost, security, speed, topology, scalability, reliability, availability

VoIP and video traffic is sensitive to delay and latency. Therefore QoS is commonly implemented to ensure these applications have enough bandwidth and are prioritised throughout the network to limit the amount of delay they incur.

Security appliances are specialised network components that typically provide many security functions, such as Cisco ASA’s and PIX, Cisco 4200 intrusion prevention and detection systems(IPS & IDS), and Cisco 3000 VPN Concentrators.

The media types used in LAN’s include copper and fibre cabling. Ethernet, Fast Ethernet(FE), Gigabit Ethernet(GE), Token Ring, and fibre distributed data interface(FDDI) are frame types used to communicate between components on fibre and copper.

Four basic types of connections or circuits are used in WAN services:

Circuit-Switched, Cell-Switched, Packet-Switched, and dedicated connections.

LAN’s provide high-speed bandwidth connections to interconnect components is geographically close locations, such as building or campus. WAN’s provide lower speed bandwidth connections to interconnect multiple locations or sites. WAN’s involve paying recurring monthly costs to a service provider.

Examples of networking devices used in WAN connections include cable and DSL modems, carrier switches, CSU/DSUs, firewalls, modems, NT1s and routers.

Network Topologies

A topology defines how the components are connected. A point-to-point topology has a single connection between two components.

In a star topology, a central device has many point-to-point connections to other components. The main problem with a star topology is that if the center of the star fails, no components can communicate with each other.

In a bus topology, all components are connected to and share a single wire.

In a ring topology, device one connects to device two, device two connects to device three, and so on to the last device, which connects back to the first device. Ring topologies can be implemented with a single or dual ring. Dual rings are for redundancy.

Fibre Distribute Data Interface(FDDI) is an example of a media technology that uses dual rings to connect computer components.

A physical topology describes how components are physically cabled together. A logical topology describes how components communicate across the physical topology. The physical and logical topologies are independent of each other.

For example, any variety of Ethernet uses a logical bus topology when components communicate, regardless of the physical layout of the cabling.

Media Type Physical Topology Logical Topology
Ethernet Bus, Star, or point-to-point Bus
FDDI Ring Ring
Token Ring Star Ring

Two types of meshed topologies are used: partial and full.

In a partially meshed environment, every device is not connected to every other device. In a fully meshed environment, every component is connected to every other component.

The formula used to determine the number of links needed to fully mesh a WAN is N x (N-1) ÷ 2, where N is the number of locations.

Network Security

Before you even consider designing a security solution for a network, you must have a sound security policy in place. The security policy will determine the security solution you will design, implement, and maintain. The Securing Cisco Network Devices course and book briefly talks about security policies. A better reference is RFC 2196 on www.ietf.org.

Four categories of common threats to networks and their components can occur: physical installations, reconnaissance attacks, access attacks, and denial of service(DoS) attacks.

Physical installations involve four types of threats: hardware, electrical, environmental, and maintenance.

Remember that most threats to a network are internal; therefore, developing a security plan to deal with physical threats is very important.

Electrical threats include irregular fluctuations in voltage, such as brownouts and voltage spikes, and complete loss of power.

Environmental threats include very low or high temperatures, moisture, electrostatic, and magnetic interference.

Maintenance threats include not having backup parts or components for critical network components; not labelling components and their cabling correctly, causing problems when performing maintenance in and around network components; and not following electrostatic discharge procedures before handling network components.

A reconnaissance attack occurs when an adversary tries to learn information about your network.

An access attack occurs when someone tries to gain unauthorised access to a component, tries to gain unauthorised access to information on a component, or increases their privileges on a network component.

DoS attacks involve an adversary reducing the level of operation or service, preventing access to, or completely crashing  a network component or service.

Designing and maintaining  networks includes factors such as cost, speed, topology, type, scalability, reliability, and availability. Certain kinds of traffic, such as VoIP and video, are sensitive to delay and latency and therefore your design should include QoS features to deal with these types of traffic. The kinds of network components you’ll see in a network include routers, switches, bridges,  hubs, and security appliances such as firewalls and IDS/IPS security appliances.