Exam Specific Topics Flashcards
(47 cards)
1
Q
What is Latency?
A
- Latency is a measure of delay in a network (ms)
- It is usually measured as round-trip delay (destination and back again)
- Particularly affects TCP/IP as delay disrupts ACK requests
- Bandwidth does not affect latency; bandwidth is a measure of data capacity, whereas latency measures the length of time taken for data to travel
- Latency affects throughput which is how much data can be transmitted by each conversation in a given time
- TCP window size determines max throughput (window size/latency)
2
Q
What is Lag?
A
- Networks contain a base amount of latency which is expected
- Lag is the sudden increase of time delay on top of latency
- Lag can have several causes:
- Internet traffic load
- Online application load
- Weather and signal interference
- Lag switches
- Overloaded router/modem
- Overloaded client device
- Malware
3
Q
What is a Collision Domain?
A
- A collision domain is a part of a network where packet collisions occur
- Collisions occur when two devices send packets simultaneously, on a shared network
- They destroy packets, so they have to be resent
- Hubs have ports that are in the same collision domain, whereas switch/router networks have separate domains for every port
4
Q
What is a Broadcast Domain?
A
- A broadcast domain is the domain in which a broadcast is forwarded
- It contains all devices that can reach each other at the data link layer (OSI L2), by using a broadcast
- All ports on a hub or switch are in the same domain by default
- All ports on a router are in different domains, and don’t forward broadcasts from one domain to another
5
Q
What is CSMA/CD?
A
- Carrier Sense Multiple Access with Collision Detection
- It helps hosts decide when to send packets on a shared network, detecting potential collisions
- It allows devices to know when a wire is being used to transmit packets
- If detected, each device will resend their packets at a random time interval
- CSMA/CD is used less regularly as switches are more common than hubs now; switches typically operate using full duplex mode, where collisions cannot happen
6
Q
What is a VLAN?
A
- Virtual Local Area Network
- It is the logical grouping of devices in the same broadcast domain
- Usually configured on switches by placing interfaces on separate domains
- Spread across multiple switches, with each VLAN being treated as its own domain/subnet
- Allows hosts which aren’t physically connected to the the same switch to be grouped together in a single broadcast domain
- Reaching hosts in a seperate VLAN requires a router
- Each port on a switch can either be an access port or trunk port
- An access port is one that can be assigned to a single VLAN
- A trunk port is one that is connected to another switch (can carry traffic from multiple VLANs)
- Frame tagging can identify which VLAN a packet belongs to (removed when forwarding to a host)
7
Q
Why are VLANs used?
A
- Increases the number of domains while decreasing their size
- Improves security by reducing number of hosts that receive copies of frames being flooded on the switch
- Hosts holding sensitive data can be kept on a separate VLAN
- Creates more flexible designs that group users by logical systems rather than physical
- Network changes are made easier by configuring ports into appropriate VLANs
8
Q
What are Loops?
A
- Looping occurs when a network has more than one active path carrying information from the same source to the same destination
- This causes the packet to amplify itself using the additional path (rather than stopping at its destination)
- This decreases network performance
- The most common cause of looping is when a device is connected to itself
9
Q
What is STP?
A
- Spanning Tree Protocol
- A Layer 2 protocol that runs on bridges and switches
- Ensures that loops aren’t created when there are redundant paths within a network
- This prevents large amounts of broadcast traffic
- It creates a spanning tree within a network of connected Layer 2 bridges
- Disables redundant links that are separate from the tree, so there is a single active path between any two nodes
10
Q
What is Subnetting?
A
- Dividing a network into two or more smaller networks
- A non-subnetted network causes problems such as:
- Single broadcast domain
- Reduced network security
- Difficult organisation
- IP address is made up of two parts: the network and the host
- Like an IP address, a subnet mask contains 32 bits
- E.g. Class A’s address contains 8 bits identifying the network, and 24 bits identifying the host
- This is because the default subnet mask of Class A is 8 bits
- Network range is determined using the binary AND operator on the IP address and subnet mask
11
Q
What is VLSM?
A
- Virtual Length Subnet Mask
- Allows you to divide an IP address space into subnets of varying sizes and host counts, without wasting many addresses
- Divides large subnets into smaller subnets
- A routing protocol that supports VLSM must be used (e.g. OSPF, RIPv2)
12
Q
What is DHCP?
A
- Dynamic Host Configuration Protocol
- Used to assign various network parameters to a device automatically
- A client-server protocol
- A client sends a request to the DHCP server, which maintains a pool of IP addresses to be assigned to requesting clients
- DHCP uses UDP port number 67 for the server and 68 for the client
13
Q
What are the four steps of a DHCP Client?
A
- Client sends DHCP-DISCOVER to discover servers
- DHCP receives DISCOVER and responds with DHCP-OFFER, which contains IP addressing information
- Client receives offers from multiple servers, but only accepts the first one; it responds with a DHCP-REQUEST for network parameters
- DHCP server approves the lease with DHCP-ACKNOWLEDGEMENT which includes lease duration among other information
14
Q
What is DNS?
A
- Domain Name System
- Translates hostnames into IP addresses
- Much more user friendly
- A DNS server has a table (dynamic or static) of hostname-to-IP address mappings
- When a user requests a resource using a hostname, a DNS request is sent to the DNS server asking for the IP address
- DNS uses UDP port 53
15
Q
What are Wildcard Masks?
A
- Used to specify a range of network addresses
- Commonly used with routing protocols and access lists
- A wildcard mask is 32 bits, and acts like an inverted subnet mask
- Zero bits indicate that the corresponding bit position must match the bit position in the IP address
- One bit indicates that the corresponding bit position doesn’t have to match the bit position in the IP address
16
Q
What is Quad Zero?
A
- Quad zero (0.0.0.0) is the default route address in IPv4
- IPv6 uses ::/0
- Any route lookup that fails to match another route will fall back to this path
17
Q
What is IPv6?
A
- Newest version of the IP protocol, developed to overcome IPv4 deficiencies (mostly, IP address exhaustion)
- Allows for 3.4 x 10^38 addresses (versus IPv4’s 4.3b)
18
Q
What features does IPv6 provide?
A
- Large (128 bit) address space
- Enhanced security using IPSec, allowing dynamic secure tunnels
- Header improvements not requiring checksum
- NAT is depreciated
- IPv6 devices can automatically configure themselves with an IP address (stateless address autoconfiguration)
19
Q
What are Public IP Addresses?
A
- Addresses reserved for use by certain entities, and not allowed to be used by private networks
20
Q
What are Private IP Addresses?
A
- Addresses which are reserved for private use behind a router (or other NAT device)
- AKA Local IP addresses
- Allow for network access without taking up public IP address space
21
Q
What are IP Address Classes?
A
- There are five classes: A, B, C, D, E
- They define a range of valid IP addresses, and the value of the first octet defines the class
- The first 3 classes are used for host addresses, whilst the other 2 are reserved for multicast and experimental purposes respectively
22
Q
What is a Routing Table?
A
- A set of rules that is used to determine where packets (travelling using an IP network) will be directed
- Contains all necessary information in order to forward a packet along the best path
- When a packet arrives, it is examined and matched to a routing table entry that provides the best match for its destination
- They can be maintained manually or dynamically
23
Q
What does a Routing Table include?
A
- Destination
- Next hop
- Interface (outgoing)
- Metric (cost to each available route)
- Routes (directly attached subnets, indirect subnets and default routes)
24
Q
What is OSPF?
A
- Open Shortest Path First
- Neighbours exchange topology information, then each OSPF router runs an SPF algorithm (often Dijkstra’s) to calculate the best routes, and adds those to the routing table
- Each router stores routing and topology in three tables: neighbour table, topology table and routing table
- Neighbours are discovered using ‘Hello’ packets
- OSPF routers go through several state changes before establishing a neighbour relationship
- Link-State Advertisements (LSAs) are used by OSPF routers to exchange topology information
- OSPF uses areas (logical grouping of networks/routers)
- All routers in the same area have the same topology table, but they are unaware of other routers
- This reduces the size of the routing table
- Routers interfacing with more than on area are called Area Border Routers (ABRs)
25
What are the main features of OSPF?
- Link state routing protocol
- Classless routing protocol
- Open standard
- Supports VLSM
- Uses areas
- Each router inside an area has the same topology
- Uses multicast addresses for routing updates
- Sends only partial routing updates
- Supports only equal cost load balancing
- Uses interface cost as a metric
- Uses SPF algorithm to calculate optimal paths
26
What are the several state changes OSPF routers go through?
- Init
- 2-way
- Exstart
- Exchange
- Loading
- Full
27
What is a Network ACL?
- Network Access Control List
- They are a set of rules used to filter network traffic
- Used on network devices with packet filtering capabilities (e.g. routers/firewalls)
- There are two types:
- - Standard access lists which filter only on the source IP address of a packet
- - Extended access lists which are more precise, and can evaluate source and destination IP addresses, Layer 3 protocol type, and source and destination ports etc.)
28
What are Routing Metrics?
- Used to make routing decisions, ultimately the best path among a list of potential routes to a destination
- Route is decided based on the lowest metric, such as hop count, speed of path, packet loss, latency
- Uses Simple Network Management Protocol (SNMP) to manage and monitor network devices and their functions
29
What are the two main methods of Managing Congestion?
- First In, First Out (FIFO): traffic is transmitted out of the interface in order of arrival
- Weighted Fair Queuing (WFQ): divides bandwidth across queues of traffic based on weights
30
What is TCP?
- Transmission Control Protocol
- A connection oriented protocol where it must first acknowledge a session between two communicating hosts
- It therefore verifies a connection before any communication takes place
- Uses a 'Three-way handshake', designed to ensure that data sent using TCP for transport is received and acknowledged
31
What is involved in the 'Three-way Handshake'?
1. A host sends the message called a SYN (Synchronise Sequence Number)
2. The receiver tells the sender a SYN ACK has been received
3. The original sender acknowledges the response of the server
32
What are the characteristics of TCP?
- Guarantees all the data is received in the correct order
| - Guarantees delivery of the data, meaning it will resend it if it doesn't arrive
33
What is UDP?
- User Datagram Protocol
- Connectionless protocol
- Does not establish a session, and doesn't guarantee delivery of data
- Less overhead means it is faster in comparison to TCP
34
What are the seven layers of the OSI Model?
```
Application (Data):
- Providing UI
Presentation (Data):
- Formats between App and Session
Session (Data):
- Establishes and terminates connections
Transport (Segments):
- Manages end-to-end data delivery
Network (Packets/Datagrams)
- Forwards data between nodes
Data Link (Frames):
- Connection between nodes
Physical (Bits):
- Converts data to bits for media
```
35
What is the OSI Model?
- Open System Interconnection
- Interconnects equipment through a standardised interface
- To integrate software and hardware, to be portable on all systems
- To creat e a model to be adhered to by all countries
- Each layer solves one of the seven problems
36
What are the benefits of the OSI Model?
- Reduced complexity (decomposed into several layers)
- Increased evolution
- Guarantees interoperability of data
- Allows modular engineering
- Standardised interfaces
37
7. Application Layer
- Provides UI for communication, such as:
| - - Email, file transfer, file management, message handling and database queries
38
6. Presentation Layer
- Formats data for exchange between application and session layer
- Masks differences of data formats between dissimilar systems
- Encodes and decodes, encrypts and decrypts, and compresses and decompresses data
- Data representation and interpretation translation is needed between two systems using different presentation standards
39
5. Session Layer
- Manages communication between applications after a connection is made; sets up the session, manages information exchanges, and then breaks it down upon the end of session
- Inter-host communication is used to set up, maintain and close down a session, providing an open communication path with another system
40
4. Transport Layer
- Manages end-to-end message delivery in the network
- Provides both reliable and sequential packet delivery through error recovery and flow control mechanisms (TCP)
- Can keep track of segments, allowing resending, and acknowledgements
- Provides unreliable transport as well, such as UDP
41
3. Network Layer
- Manages the transfer of packets that are to be forwarded on to different networks, and between nodes on the same network
- Data is routed through a network and can also be routed through interconnected networks; splitting data for transmission and re-assembling upon reception
- The IP part of TCP/IP operates at this level
42
2. Data Link Layer
- Provides a reliable link between two directly connected nodes
- Ensures that all data has been received thereby providing flow, error control and synchronisation of the physical layer
- Error detection, with the use of CSMA/CD
- Point-to-Point (PPP) is an example of the data link layer in the TCP/IP stack
- Comprises two layers: Logical Link Control Layer (LLC) and Media Access Control Layer (MAC)
43
1. Physical Layer
- Transmission of binary data via a medium
- - It defines the cabling, or method of data carriage, connectors, electrical characteristics of the channel and the signal
- - It also defines the protocol to establish and terminate a connection between two nodes
- The NIC converts the binary data into electrical voltages
- Devices operating at this layer include: network cards, hubs, bridges, switches and routers
44
What does the LLC Layer do?
- Provides flow control, acknowledgement and error notification
- Controls data exchanged between the source and destination machines
45
What does the MAC Layer do?
- Determines who is allowed to access the media at any one time, using CSMA/CD
- Determines where one frame ends and another starts (frame synchronisation)
46
What is Circuit Switching?
- Establishes a point to point communication between two devices
- Channel is dedicated
- Full bandwidth given
- Remains connected for duration
- Designed for voice traffic
- Highly inefficient
- Minimal delays
47
What is Packet Switching?
- More efficient than circuit switching
- Data segmented into small chunks, called packets
- - Each has a destination and a source address
- People can communicate at the same time
- No dedicated link
- Maximises bandwidth
- Checks on packet validity, using checksum and CRC
