Block 3 unit 1-6

Hub

Simple device that copies data and sends it to every device connected to it; Legacy Device

Switches

Uses ASICs to help path Determination.

Learn MAC address of each connected devices

Break up Collision Domains

Switches are fast and have adjustable speed/bandwidth.

Switches Segment networks

PPP

Peer-to-peer; Connecting directly to another device without the need of a network

STP or IEEE 802.1D

Spanning tree protocol; Loop Prevention Protocol

Ensures efficient and reliable data transmission

UDLD

Detects traffic flowing in direction due to bad TX/RX and shuts down the port

CDP

Cisco Discovery Protocol; Stores MAC, VLAN, Management IP, system names etc.

Similar to LLDP

LLDP (Link Layer Discovery Protocol)

Non-proprietary protocol that allows neighboring devices to learn abut each other

Stores MAC, VLAN, Management IP, System name, etc

802.1q

Non-proprietary trunking Protocol

ISL

Cisco proprietary VLAN trunking protocol

Switching Methods

Store-and-forward, Cut-through, and Fragment-free

Store-and-forward

Stores entire frame into memory and checks it for errors using CRC (Cyclic Redundancy Check)

Slow but high level of error-free traffic

Cut-Through

Copies destination MAC to memory and sends the frame through

Used for speed

Fragment-Free

Hybrid of Cut-through & Store-and-Forward

Checks 64 bytes before sending it off

Most errors occur in the first 64 Bytes

Loops

Multiple paths to the same destination

Can cause, Broadcast Storms, Multiple frame copies, and Multiple Loops

Broadcast Storms

Network devices rebroadcasting data to the network

Multiple frame copies

Device receives multiple frame copies from different LAN segments

Multiple loops

Loops can occur within loops, consuming bandwidth.

Spanning tree protocol IEEE 802.1D is used to prevent loops

VLAN

Logical (or IP) grouping of Network devices

Layer 3 Routing Process

VLAN benefits

Increased Security

Flexibility and Scalability

Removes geographical barriers

VLAN 1 by default

VLANS are made with a number and Description (Ex: VLAN 10 is named Sales)

Trunk Port AKA Trunk Link

Connection that carries multiple VLANs; Having Multiple different VLANS on one cable

Inter-Switch Link (ISL)

Cisco Proprietary VLAN Trunking

IEEE 802.1Q

Open Standard VLAN trunking

Management VLAN

Used for Telnet, SSH, SNMP, and Syslog

Should NEVER be assigned to VLAN 1

Router

Connects 2 or more lines from different networks. These form internetwork.

Routing

Sending a packet from one device to another one a different network

Multilayer switches (MLS)

Can operate at layer 2 and 3 of OSI Model.

Sometimes referred as a Layer 3 Switch

External Components of a switch

Console Port, Auxiliary port, network interfaces

Console Port

Allows local configuration of a Switch

Auxiliary Port

Allows configuration of a switch via a Modem. Is often disabled due to security concerns

Internal Components of a Switch

ROM, Flash Memory, NVRAM, and RAM

ROM

Read only memory; Diagnostic and Boot-up Routines

Flash Memory

Stores the OS

NVRAM

Stores the startup Configuration and loads it into RAM

RAM

Running configuration; Used for operations

Gateways

Router that sends IP packets between networks/subnets

Autonomous System (AS)

Defined by RFC 1930; A network that works without human interference (Most networks)

Types of Gateways

Interior Gateway, Exterior Gateway, and Default Gateway.

Interior Gateway

Routers Moving Information WITHIN a Network

Exterior Gateway

Routers Moving Information BETWEEN networks

Default Gateway

Access Point to a network

Networking Challenges

Connectivity, reliable service, Network Management, and Flexibility

Connectivity

Interconnected devices, media, and speeds

Reliable Service

Up-Time; consistent access to a network

Network Management

Troubleshooting capabilities/Centralized support

Flexibility

Expansions, upgrades, and new apps/services

Network Segmentation

Routers and switches are used to segment networks.

Routers break up broadcast domains.

L3 Broadcast

IP prefix for a subnet/network

Protocols

Rules that govern how devices communicate

Path Determination

Routers build & Maintain routing tables using Routing Protocols

Packet Switching

Sending data (Or packets) from router to router

As the data travels, it's destination MAC changes.

The IP source and Destination stay the same

Conditions in which a packet is discarded

TTL (Time-to-live) Expires

Version number of packet header is wrong

Error in Transmission

Packet is fragmented and the fragmentation flag isn't set

ROUTED protocols

Deliver the messages

IP, ARP, Hello Protocol, & ICMP

Hello Protocol AKA Neighbor Discovery

Enables network devices to learn about other network devices.

ICMP - Internet Control Message Protocol

Reports errors with IP packet processing

ROUTING protocols

Responsible for Path Determination & Packet Switching.

Build and Maintain Routing Tables.

RIP, EIGRP, OSPF, BGP

Different types of Routes

Static, Default, and Dynamic

Static Routes

Manually configured by System Administrator

AD=1 is the most trusted (Because there's nothing higher than 1 in Binary)

Dynamic Routes

Routers use information to build and main routing tables;

Finds the most efficient way to reach it's destination

Default Routes

Gateways of Last Resort

Type of Static route

Used by router if no other path is known to the destination

Path Length

Most relied upon metric

Hop count

number of internetworking devices from source to destination

Metrics to determine best route

Reliability, Delay, Bandwidth, and Load

Delay

How long it takes the packet to move to the destination

Bandwidth

Available traffic capacity of a link

Load

How much traffic is on the network/How busy it is

Distance Vector Algorithms

Algorithms that call for each router to send it's entire routing table (Similar to Hello protocol and CDP)

Advantages of Distance Vector Algorithms

Simple to understand and config.

Less processor demand.

Cheaper

Disadvantages of Distance Vector Algorithms

Limited by hop count (Max of 15)

Doesn't support VLSM (Variable Length Subnet Masking)

Builds routing table from neighbors (Explanation: they could be wrong)

Link State Algorithms

Also called Shortest Path First (SPF)

Builds Topological map of the area into memory.

Interior Gateway Protocols

RIP (limited to 15 hops & is used in Distance Vector)

OSPF (Open Shortest Path First, is used in Link State)

EIGRP (Enhanced Interior Gateway Routing Protocol)

Advantages of OSPF

Supports large networks

Fast convergence

No routing loops

Supports classless routing

Smaller updates

Disadvantages of OSPF

Complex topology

Needs highly trained staff

Takes up allot of memory

SPF algorithms use more processing power

EIGRP

Metrics: Bandwidth & delay

Known as “Balanced Hybrid” because it combines advantages

Resembles both Distance Vector and Link State Algorithms

Advantages of EIGRP

Fast convergence when feasible successor is available

Smaller updates

Scales to large networks

VLSM supported

Disadvantages of EIGRP

High memory (AKA money) requirements

Routing algorithm is complex

No area concept

BGP

BETWEEN AS (Autonomous System)

Loop free

Doesn't use metrics like Delay, hop count, or bandwidth instead is based off of network policies set by administrators

Core layer

Core of the network. Also called High Speed Backbone.

Transmits large amounts of data quickly and reliably.

Core Layer Failure causes

Faulty Link or device

Core Layer Protocols

Uses routing protocols with low convergence times (Fast & efficient Layer 2 switching devices)

Distribution Layer

Communication point between access layer and core layer. Provides routing, filtering, WAN access, & determines how many packets access the core layer.

Determines fastest way to send data

Distribution Layer Policies

Uses tools likes: Access lists, packet filtering, for , security, and network policies.

Routes betweens VLANs.

Aggregation point for multiple access sileitches and must be capable of handling large amount of traffic

Client Connection

Connects multiple desktop devices to the Distribution Layer

Many access points and low cost

Shared & switched bandwidth devices

Filters based on MAC

Subnet

Network within a network

Logic division of a network of connected devices based on IP

Subnet Process

3 parts of an IP, Network (1-126 and 128-228), Subnet Mask, and Host Portion (0-255)

Subnetting: Networking portion borrows bits from the host portion

Subnetting rules

Beginning of Default mask and class cannot change

Class B network and cannot be 255.0.0.0 or 255.255.255.255

Classes must start with default mask, however subnet mask may change if it follows previous rules

Benefits of subnetting

Reduced network traffic

Optimized performance

Simplified management

Class A Subnetting

Up to 8,388,608 available subnets per class A address.

Takes in Host requirements when Subnetting

Class A provides most available subnets

Class B subnetting

Up to 32,768 subnets

Also takes in Host requirements when Subnetting

Class C Subnetting

Used for small networks (Just like Class C networks)

Least amount of borrowable bits

Classless Inter-Domain Routing (CIDR)

Written with IP address followed by a / and the subnet number: Example 192.168.12.10/27

Variable Length Subnet Masking (VLSM)

Used for Subnetting a network

Subnets can be broken into further subnets

Allows more efficient use of IP addresses

Topological map

Visual Representation of a network

Can be physical or logical

Bus topology

All nodes (aka devices) connects by central medium (or bus) which has exactly 2 endpoints

Bus Topology benefits & Disadvantages

Easy to configure

Requires less cable length

Disadvantages

Hard to pinpoint a problem

If the central bus goes down, so does the network.

Star Topology

One central hub/switch which all data passes through to reach each device

Most common topologies

Very reliable due to individual devices not affecting the network if they crash.

Disadvantages

Higher cable costs

If central hub/switch fails, no connected devices can access it

Tree topology

Hybrid of Bus & Star Topology

Mesh Topology: Full Mesh

Every node is directly connected to every other node (Easy to make loops)

Mesh Topology: Partial Mesh

Nodes are only connected to the nodes they interact with the most