Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Year 3 > Networking > Flashcards

Networking Flashcards

1
Q

OSI Physical Layer

A
  • Is responsible for the transmission and reception of bits between a device and a physical transmission medium
  • It converts these digital bits into electrical, radio, or optical signals
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

OSI Data Link Layer

A
  • Provides node-to-node data transfer (a link between two directly connected nodes)
  • It detects and possibly corrects any transmission errors that may occur in the physical layer
  • Is divided into two sublayers: MAC and LLC
  • MAC (media access control) is responsible for controlling how devices in a network gain access to a medium and permission to transmit data
  • LLC (logical link control) is responsible for identifying and encapsulating network layer protocols, and controls error checking and frame synchronization
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

OSI Network Layer

A
  • Provides the means of transferring variable length packets from one node to another connected in a different network
  • It also manages data transfer rates, via flow control as a means of easing congestion
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

OSI Transport Layer

A
  • Deals with the packetisation of data that comes in from the session layer so that it’s suitable for the network layer
  • Provides the acknowledgement of the successful data transmission and sends the next data if no errors occurred
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

OSI Session Layer

A
  • Manages connections between source and destination
  • This layer is responsible for responsible for gracefully closing a session as well as session checkpointing and recovery in the case of interruptions
  • Allows the establishment, use and termination of a connection
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

OSI Presentation Layer

A
  • Makes decisions on how data is represented before it is sent across the network
  • Takes care of data compression and encryption/decryption
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

OSI Application Layer

A
  • The layer closest to the end user
  • This layer and the user interact directly with the software application
  • This layer also serves as a window for the application services to access the network and for displaying the received information to the user
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

TCP/IP Link Layer

A
  • Corresponds to the OSI physical and data link layers
  • Ensures that the model can send and receive data from other layers
  • Defines the protocols and hardware required to connect a host to a physical network and to deliver data across it
  • Packets from the Internet layer are sent down this layer for delivery within the physical network
  • The destination can be another host in the network, itself, or a router for further forwarding
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

TCP/IP Network Layer

A
  • Corresponds to the OSI network layer
  • Performs two functions:
    1) Host addressing and identification: this is accomplished with a hierarchical IP addressing system
    2) Packet routing: this is the basic task of sending packets of data (datagrams) from source to destination by forwarding them to the next network router closer to the final destination
  • It also defines the protocols which are responsible for logical transmission of data over the entire network: IP, ICMP, ARP
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

TCP/IP Transport Layer

A
  • Corresponds to the OSI transport layer
  • Is responsible for end-to-end communication and ensuring delivery of data is error-free
  • Also shields the upper-layer applications from the complexities of data
  • Contains two protocols: UDP and TCP
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

TCP/IP Application Layer

A
  • Corresponds to the OSI session, presentation and application layers
  • Is responsible for node-to-node communication and controls user-interface specifications
  • Contains the communications protocols and interface methods used in process-to-process communications across an IP network
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

CSMA/CD

A
  • Forces the sender to send transmissions to check for the presence of a digital signal on the wire
  • If no other hosts are transmitting packets, the sender begins sending the frame
  • The sender also monitors the wire to make sure no other hosts begin transmitting
  • However, if another host begins transmitting at the same time and a collision occur, the transmitting host sends a jam signal that causes all hosts on the network segment to stop sending data
  • The CSMA/CD rules define how long the device should wait if a collision occurs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Metcalfe’s Law

A

A network’s value increases exponentially with the size of the network (in this case, the users)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

IP address types

A
  • Unicast (a single packet destination)
  • Broadcast (one packet goes to every local host)
  • Multicast (one packet goes to one or more hosts)
  • Anycast (IPv6 only, a packet goes to any one of a selection of servers, usually the closest in some sense)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

IP routing classes

A
  • Local hosting, requiring an interior gateway protocol
  • Non-local hosting, requiring an exterior gateway protocol
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Classful systems

A
  • Original architecture used until CIDR was introduced
  • Divides IP addresses into one of five classes, depending on its four leading bits
  • Class A:
    1) Most significant bit sequence is 0
    2) Next 7 bits gives network number
    3) Leaves 24 bits for determining the host in any of the networks
  • Class B:
    1) Most significant bit sequence is 10
    2) Next 14 bits gives network number
    3) Leaves 16 bits for determining the host in any of the networks
  • Class C:
    1) Most significant bit sequence is 110
    2) Next 21 bits gives network number
    3) Leaves 8 bits for determining the host in any of the networks
  • Class D:
    1) Most significant bit sequence is 1110
    2) Next 28 bits gives network number
    3) Used for multicasting (when one host sends data to more than one destination)
  • Class E:
    1) Most significant bit sequence is 11110
    2) Next 27 bits gives network number
    3) Reversed for future use
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Motivation for introducing classful addressing

A
  • Needed to deal with the expansion of the Internet, since the previous header could only name 256 networks
  • This expansion was to cater for the trend of increasing local area networking
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

CIDR

A
  • Stands for Classless Interdomain Routing
  • Unlike classful routing, CIDR allows for blocks of IP addresses (typically class C) to be allocated by region and then allocated to ISPs who then assigned addresses to their customers
  • CIDR has been effective in making good reuse of class A networks
  • It is subnet mask-based, meaning that the IP address is split into a network and host address, with focus on the host address
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Why CIDR?

A
  • IP addressing used to be class-based, and allocations were based on the bit boundaries of the four octets of an IP addresss:
    1) Class B networks were limited to 65535 network interfaces
    2) Class C networks were limited to 254 network interfaces
  • This led to inefficiencies as well as exhaustion of IP address availability a large numbers of class C networks were required with individual route announcements, leading to exhaustion of
  • As a result, the system of allocating IP addresses was no longer scalable
  • Since CIDR is VLSM-based, it allows the division of a network into arbitrarily sized subnets
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

TCP header flags

A
  • URG: urgent data
  • ACK: acknoledgement field
  • PSH: push this data to the application as possible
  • RST: reset the connection
  • SYN: synchronise a new connection
  • FIN: finish a connection
  • ECE: congestion notification
  • CWR: congestion window reduced
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

SYN

A
  • Only the first packet sent from each end should have this flag set
  • Used as the first step in establishing a 3-way handshake between hosts
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

ACK

A
  • Indicates that the acknowledgement field is significant
  • Acknowledges that a packet has been successfully received
  • In the second step of the 3-way handshake, the receiver sends an ACK = 1 along with a SYN = 1 to notify the sender that it received the initial packet
  • All packets after the initial SYN packet sent by the client should have a set ACK flag
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

FIN

A
  • Indicates the last packet from the sender
  • Is used to request for connection termination (when there is no more data from the sender)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

RST

A
  • Resets the connection
  • Can be used to terminate the conection if the RST sender thinks there is an issue with the TCP conection
  • Usually gets sent to the sender from the receiver when a packet is sent to a particular host that was not expecting it
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

URG

A
  • Indicates that the urgent field is significant
  • Data with this flag is for
  • Notifies the receiver to process the urgent packet before processing all other packets
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

ECE

A
  • Depends on the SYN flag value, this flag checks if a TCP peer is ECN capable during a 3-way handshake
  • If SYN flag = 0, indicates that there is network congestion to the TCP sender
  • If SYN flag = 1, then peer is ECN capable
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

CWR

A
  • Set by the sending host to indicate it received a TCP segment with the ECE flag set and had responded in congestion control mechanism
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

PSH

A
  • All data in the buffer is pushed to the receiving application
  • Indicates that the data should be passed as soon as possible
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

TCP reliability

A
  • If one host sends a packet to another, the receiving host needs to send an ACK packet to inform the sender that it has received it
  • Otherwise, the sender just resends the packet
  • Every byte in a TCP connection is numbered
  • The sequence number is initialised to some random number and represents the byte number of the first byte of data in the TCP packet sent
  • The acknowledgement number is the sequence number of the next byte the receiver expects to receive
  • The receiving host ACKs the client’s sequence number by incrementing it by 1
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

Three-way handshake

A
  • Used to open a TCP connection
  • There are three steps:
    1) SYN: The active open is performed by the client sending a SYN to the server. The client sets the segment’s sequence number to a random value A.
    2) SYN-ACK: In response, the server replies with a SYN-ACK. The ACK number is set to one more than the received sequence number, and the sequence number that the server chooses for the packet is another random number, B.
    3) ACK: Finally, the client sends an ACK back to the server. The sequence number is set to the received ACK value, and the ACK number is set to one more than the received sequence number
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

Passive open

A

The server must first bind to and listen at a port to open it up for connections before a client attempts to connect with a server, occurs once the active open is established

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

Active open

A
  • The creation of a listening socket, to accept incoming connections
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

Active close

A
  • The side that initiates the closedown process by issuing the first close() call is said to initiate an active close
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

Passive close

A
  • The side that closes in response to the initiation is said to initiate a passive close
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

TCP segment exchange

A
  • Uses up to four segments
  • When an endpoint wishes to stop its half of the connection, it transmits a FIN packet, which the other end acknowledges with an ACK
  • After the side that sent the first FIN has responded with the final ACK, it waits for a timeout before finally closing the connection
  • A connection can be “half closed”, in which case one side has terminated its end, but the other has not
  • Therefore the side that has terminated can no longer send any data into the connection, but the other side can
  • The terminating side should continue reading the data until the other side terminates as well
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

TCP three way handshake termination

A
  • A second way of terminating TCP connections
  • Merely combines two steps into one, when a host sends a FIN and the receiving host replies with a FIN & ACK and the host that sent the original FIN replies with an ACK
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

TCP reset sending termination

A
  • A third way of terminating TCP connections
  • Only used in the case of errors, such as connection to a non-existent port
  • When a host receives a RST packet, it ends the connection and discards any packets in transit
  • There are no ACKs required, therefore the connection ends immediately
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

TCP flow control

A
  • Adjusts how frequently packets are sent depending on the network conditions and the state of the host receiving these packets, in order to ensure reliability
  • The advertised window manages the state of the receiving host
  • Different from congestion control in that flow control prevents the end-node from being overwhelmed
  • Uses 16 bits of window size
  • The destination has a limited amount of buffer space and fills up if data is not being processed quickly enough
  • After receiving a segment, the receiving window sends back to the sender a segment indicating the amount of space left
  • That way, the sender can slow down the rate of transfer for the buffer to free up space
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

Congestion window

A
  • A TCP state variable that determines the amount of data the TCP can send into the network before receiving an ACK
  • When a connection is set up, the congestion window is set to a small multiple of the MSS allowed on that connection
  • It is calculated by estimating how much congestion there is on the link
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

Advertised window

A
  • An adjustable field that determines the speed of transferring packets to make best use of current conditions in the receiving host
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

Sliding window

A
  • Describes the range of bytes that the sender can send at one time
  • The sender recomputes the availability of the buffer after every ACK it receives
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
42
Q

Describing the sliding window

A
  • The window size is sent in every ACK segment
  • There are two edges on the sliding window
    1) The left hand edge is defined by the number of the latest ACK segment
    2) The right hand edge is defined by adding on the window size that is included in the ACK segment
  • The sliding window closes as more ACKs are received and the left hand edge advances (moves further left)
  • The window opens as the application reads data and the right hand edge advances (moves further right)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
43
Q

Window scale field

A
  • An optional field that can increase the receive window size allowed in TCP above its former maximum value of 65,535 bytes
  • Is necessary because modern networks need to get the most out of the available bandwidth, which can only be achieved with a large window
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
44
Q

Delayed ACKs

A
  • Used to reduce traffic
  • A host may delay sending an ACK response by up to 500 ms
  • Additionally, with a stream of full-sized incoming segments, ACK responses must be sent for every second segment
  • Delayed ACKs can give the application the opportunity to update the TCP receive window and also possibly to send an immediate response along with the ACK attached
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
45
Q

Piggybacking

A
  • Wherever a frame is received, the receiver waits and does not send the control frame (ACK) back to the sender immediately
  • The receiver waits until its network layer passes in the next data packet
  • The delayed acknowledgement is then attached to this outgoing data frame
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
46
Q

Slow start

A
  • The congestion window is initialised to the destination’s MSS
  • A threshold value, ssthresh, is initialised to 64kb
  • The value for the congestion window size increased by one with each ACK received, effectively doubling the window size each RTT
  • The increase in tranmission rate continues until either a loss is detected or ssthresh is reached
  • If a loss event occurs, TCP assumes that it is due to network congestion and takes steps to reduce the offered load on the network
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
47
Q

Congestion avoidance

A
  • Starts when ssthresh is reached without any problems
  • At this point, the window is increased by 1 segment for each round-trip delay time (RTT), making growth linear rather than near-exponential like slow start
  • The window continues to increase until the network’s limit is reached, usually dute to a timeout
  • When this happens, TCP assumes this is due to network congestion
  • The following steps occur:
    1) Congestion window is reset to MSS = 1
    2) ssthresh is set to half the congestion window size before the timeout
    3) The congestion window starts increasing again, irrespective of whether slow start was running again
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
48
Q

Fast Retransmit

A
  • Reduces the time a sender waits before retransmitting a lost segment
  • After receiving a packet an acknowledgement is sent for the last in order byte of data received
  • When a sender receives three duplicate acknowledgements, it can believe that the segment carrying the data that followed the last in order byte specified in the acknowledgment was lost
  • A sender with fast retransmit will then retransmit this packet immediately without waiting for its timeout
  • On receipt of the re-transmitted segment, the receiver can acknowledge the last in order byte of data received
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
49
Q

NAT

A
  • Stands for Network Address Translation
  • A method of mapping many local IP addresses into one global IP address to provide Internet access to the local hosts, packet addresses are modified as they traverse the gateway
  • This works by allowing a single device, usually a router, to act as a gateway, meaning that only a single unique IP address is required to represent an entire group of computers to anything outside their network
  • This helps in mitigating address exhaustion
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
50
Q

Deployment of IPv6

A
  • Considered the ultimate solution to address exhaustion since it’s a completely new version of IP
  • It was designed to have a significantly larger address space (approximately 2¹²⁸ addresses) and more simplified headers to improve processing of packets
  • Routers will never fragment as packets are dropped and error messages are sent to the sender (every IPv6 host is required to use path MTU discovery to avoid fragmentation)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
51
Q

Solving address exhaustion

A
  • The entire range of usable IPv4 addresses has now been depleted, despite the mitigating efforts of CIDR and NAT
  • CIDR has simply extended the time before exhaustion by around 3 years
  • As a result, IPv6 is being developed and most ISPs are deploying it

Issues with CIDR:
- It is no longer possible to determine by looking at the first octet to determine how many bits of an IP address represent the network ID and how many the host ID

Issues with NAT:

  • In the private network, the NAT becomes the endpoint meaning that the device knows only its private IP address, which can’t be accessed from the internet (effectively breaking the end-to-end principle, which is key in the development of new applications)
  • An issue of compliance, developers have to use port 80 (the HTTP port) to work around NATs. However, reuse of this port makes applications process traffic through that port, which makes said traffic appear as basic web browsing, which is infringing.

Issues with IPv6 deployment:

  • Cost: it’s expensive as several software upgrades may be required, since the majority of systems won’t have IPv6 support
  • Lack of demand: consumers don’t care about the protocol, just accessibility to content and services, therefore there is no ability for ISPs to charge for them
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
52
Q

UDP

A
  • Stands for User Datagram Protocol
  • If a node wants to send UDP data, it first creates a socket, then sends the data to that socket
  • If a node wants to receive UDP data, it first creates a socket on an address that is known by the node that will send the data, then reads the data from that socket
  • UDP layers are thinner than TCP
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
53
Q

WPA-PSK

A
  • Works by firstly configuring a password, between 8 and 63 characters
  • This password derives a key, depending on the access point
  • Using this key, the host device can then be authenticated in the access point
  • Every device on the network shares the same password and as a result, share the same key
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
54
Q

WPA Enterprise

A
  • Each user self-authenticates via a server (usually a RADIUS one)
  • RADIUS server authenticates users by account certificates
  • Each device each has its own password
  • Works by assigning a long encryption key to each connected device
  • The key is not visible as it’s only created when a user presents their login credentialst
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
55
Q

Wi-Fi Protected Setup

A
  • A wireless network standard that attempts to make it easier for wireless devices to connect to a router
  • Only works with networks that are encrypted with a WPA-PSK key
  • Works by two ways:
    1) Pressing a dedicated WPS button on the router and then on the device, which allows the device to connect without the need for a password
    2) Using an eight-digit pin that is generated by the router
  • However, the pin is easy to hack via brute force attack
  • This is because the pin is stored in two blocks of four digits, thus the first block can be brute-forced (there would only be 10000 possiblilities) and the second block can be done the same way
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
56
Q

Carrier Sense

A
  • The wire is listened to in order to determine whether there is a signal passing along it
  • A frame cannot be transmitted if the wire is in use
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
57
Q

Multiple access

A
  • Allows multiple devices to share the same amount of wire to transfer data between themselves
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
58
Q

CSMA/CA

A
  • Stands for carrier sense multiple access with collision avoidance
  • Used in WiFi
    1) In wireless networks, there is no way for the sender to detect collisions the same way CSMA/CD does since the sender is only able to transmit and receive packets on the medium but is not able to sense data traversing that medium
    2) Should the control message collide with another control message from another node, it means that the medium is not available for transmission and the back-off algorithm needs to be applied before attempting retransmission
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
59
Q

Hidden host problem

A
  • A situation where a host can communicate with a central host, but can’t communicate with other nodes that are communicating with that central host
  • Therefore, collisions can’t be detected between hosts that transmit to the central node
  • Can be addressed using RTS/CTS
60
Q

RTS/CTS

A
  • Stands for Request to Send/Clear to Send
  • When a host wants to transmit data to another host, it sends out an RTS packet
  • The receiving host then replies with a CTS packet
  • Both the RTS and CTS pakets contain a duration field, which designates a time period for the transmission to occur
  • Every other host will see the CTS packet, therefore will wait to send anything until after the transmission completes
  • The ACK flag of the receiving host is then set to visible
  • Other hosts can now send data again
  • Is optional because is not required in every situation
  • Problem: extra latency
61
Q

MTU

A
  • Stands for Maximum Transmission Unit
  • Is the size of the largest packet that can be transferred in a single network layer transaction
62
Q

Path MTU

A
  • The smallest MTU for the entire path travelling from source to destination

IPv4:
- A datagram not larger than the path MTU will not get fragmented
- MTU Discovery works by sending variously sized datagrams with DF set, and monitors the errors returned
- When a datagram reaches the destination with no fragmentation error we have found a lower bound for the path MTU
- This bound is approximate as the network is dynamic and paths may change
- We can send segments of decreasing size, starting with the minimum of the MSS of the sender and the MSS announced by the other end, or 536 if the other end did not give an MSS and with the IP flag DF set
- If an ICMP error “fragmentation needed but DF set” happens during a TCP connection, the congestion window should stay as it is and initiate a slow start
- It’s important to allow ACKs to come through again so the window backs off
IPv6:
- Does not support fragmentation, and as a result doesn’t support DF
- Path MTU discovery is required
- Any device along the path whose MTU is smaller than the packet will drop the packet and send back an ICMPv6 Type 2 message, allowing the source host to reduce its Path MTU appropriately
- The process is repeated until the MTU is small enough to traverse the entire path without fragmentation

63
Q

Issues with path MTU discovery

A
  • A router can generate and send an ICMP message, but the ICMP message gets blocked by a router or firewall between this router and the sender
  • A router can drop a packet and not send an ICMP message (not very common)
  • A router can generate and send an ICMP message, but the sender ignores the message (not common)
64
Q

Fragmentation

A
  • The breaking of packets into smaller pieces
  • This is done so that the resulting pieces can pass through a link with a smaller MTU than the original packet size
  • Otherwise it is discarded
65
Q

IPv4 header fields

A
  • Version (4 bits)
  • Header length (4 bits)
  • Type of service (8 bits)
  • Total length (16 bits)
  • Identification (16 bits)
  • Flags (3 bits)
  • Fragment offset (13 bits)
  • Time to live (8 bits)
  • Protocol (8 bits)
  • Checksum (16 bits)
  • Source address (32 bits)
  • Destination address (32 bits)
  • Header options (Varies)
  • Data
66
Q

Issues with IP fragmentation

A
  • There is a small increase in CPU and memory overhead in order to fragment an IPv4 datagram
  • If one fragment of an IPv4 datagram is dropped, then the entire original IPv4 datagram must be resent, and it is also fragmented
67
Q

IPv4 Version Field

A
  • Contains the number 4, representing in 4 bits (0100) and is needed to know how to interpret different packets
68
Q

IPv4 Header Length Field

A
  • Contains the length of the IP header
  • The bits specify the number of 32-bit words in the header
  • Is required because the end of the header needs to be distinguished from the optional fields
69
Q

IPv4 Type of Service Field

A
  • Specifies to routers how the datagram should be handled and request a route that has optimal throughput and reliability
  • Bits 5,6,7 are labelled as “Precedence”, denoting the priority of the datagram
  • The second field, labeled “TOS”, denotes how the network should compromises between throughput, delay and reliability
  • The first field is unused and and set to zero
  • Usually used for QoS purposes
70
Q

IPv4 Total Length Field

A
  • Define the entire packet size in bytes, including header and data
  • The minimum size is 20 bytes (header without data), and the maximum is 65535 bytes
  • Is required so the machine knows when the current packet ends so it can start processing the next
71
Q

IPv4 Identification Field

A
  • Used to uniquely identify a group of fragments of a single IP datagram for a given source address, destination and protocol
  • Works in support of fragmentation and reassembly
  • When an IP datagram is fragmented:
    1) Each fragmented datagram is assigned the same identification number
    2) This ID number is useful when reassembling fragmented datagrams
    3) It helps to identify which IP datagram, the fragmented datagram belongs to
72
Q

IPv4 Flag Fields

A

Bit 0: Reserved, has to be zero
Bit 1: DF (Don’t fragment)
Bit 2: MF (More fragment)
- These flags are used to control or identify fragments

  • If DF is set to zero, it grants permission to intermediate devices to fragment the datagram if needed
  • If DF is set to one:
    1) It indicates the intermediate devices not to fragment the IP datagram
    2) If network requires the datagram to be fragmented to travel further but settings does not allow its fragmentation, then it is discarded
    3) An error message is sent to the sender saying that the datagram has been discarded
  • If MF is set to zero, it tells the receiver that the current datagram is either the last fragment in the set or that it is the only fragment
  • If MF is set to one:
    1) It tells the receiver that the current datagram is a fragment of some larger datagram
    2) More fragments follow
    3) The MF bit on every fragment except the final one is set to 1
73
Q

IPv4 Fragment Offset Field

A
  • Specifies the offset of a particular fragment relative to the beginning of the original unfragmented IP datagram
  • The first datagram has an offset of zero, which allows a maximum offset of 65528 bytes (calculated by (2¹³ - 1) x 8), which exceeds the maximum IP packet length if the header length is included (the offset would then be 65548 bytes)
74
Q

IPv4 Time to Live Field

A
  • Helps prevent datagrams from looping around during routing
  • Indicates the maximum number of hops a datagram can take to reach the destination
  • The value of TTL is decremented by 1 when:
    1) The datagram takes a hop towards the destination
    2) The datagram takes a hop to any intermediate device (such as a router)
  • If the TTL value becomes zero, the datagram is discarded and an ICMP Time Exceeded message is sent back to the source address
75
Q

IPv4 Protocol Field

A
  • Tells the network layer at the destination host which protocol the IP datagram belongs to
  • Protocol number of TCP is 6 and UDP is 17
76
Q

IPv4 Checksum Field

A
  • Detects corruption in the header of IPv4 packets
  • If there is no corruption, the result of summing the entire IP header, including checksum, should be zero
77
Q

IPv4 Source Address Field

A
  • The IPv4 address of the sender of the packet
78
Q

IPv4 Destination Address Field

A
  • The IPv4 address of the receiver of the packet
79
Q

IPv4 Header Options Field

A
  • Allows for extensions that are not present in the original datagram specification, such as security
  • They are not used often
  • They must be considered if the header length is more than 5 bits
80
Q

IPv6 Header Fields

A
  • Version (4-bit)
  • Traffic class (8 bits)
  • Flow label (20 bits)
  • Payload length (16 bits)
  • Next header (8 bits)
  • Hop limit (8 bits)
  • Source address (128 bits)
  • Destination address (128 bits)
81
Q

IPv6 Version Field

A
  • Contains the number 6
  • Same size as its IPv4 equivalent but has limited use,
82
Q

IPv6 Traffic Class Field

A
  • Indicates the class or priority of the IPv6 packet
  • Holds two values
  • The 6 most significant bits hold the DS field, which is used to classify packets
  • The remaining 2 bits are used for ECN
83
Q

IPv6 Flow Label Field

A
  • Allows routers to recognise packets in a single flow and treat them identically
  • Flow = group of packets
  • If the flow label = 0, then the packet does not belong to any flow
  • For default router handling, the flow label is set to zero
84
Q

IPv6 Payload Length Field

A
  • Contains the length of the data field in octets, including any extension headers and the upper layer of the PDU (Protocol Data Unit)
  • For payload lengths greater than 65535 bytes, the field is set to zero and a “jumbo payload” option is used
  • Needed to identify when the packets starts and ends
85
Q

IPv6 Next Header Field

A
  • Indicates type of the first extension headers if present, immediately following the IP6 header, or indicates the protocol in the upper layer PDU
86
Q

IPv6 Hop Limit Field

A
  • Decrements by one, by each node that forwards a data packet
  • If the field is decremented to zero, then any remaining packets are discarded
  • The main purpose of this field is to identify and discard packets that are stuck in an infinite loop as a result of routing errors
87
Q

IPv6 Source Address Field

A
  • Contains the address of the original source of the packet
  • Four times larger than its IPv4 equivalent
88
Q

IPv6 Destination Address Field

A
  • Contains the address of the final destination of the packet
  • Four times larger than its IPv4 equivalent
89
Q

IPv4 vs IPv6

A
  • Version
    1) Is in IPv6 but with a different number (6 instead of 4)
  • Header Length
    1) Is removed in IPv6
    2) IPv6 does not have a header length field because the IPv6 header is always a fixed length of 40 bytes
    3) IPv4 has a header option field therefore actual header size varies
    4) In IPv6, each extension header is either a fixed length or indicates its own length
  • Type of Service
    1) Is replaced by the IPv6 Traffic Class field
  • Total Length
    1) Is replaced by the IPv6 Payload Length field, which indicates only the size of the payload
  • Identification, flags and fragment offset fields
    1) Are removed in IPv6
    2) Fragmentation information is not included in the IPv6 header
    3) This info is instead contained in a fragment extension header
  • Source and destination address
    1) These fields are the same except that IPv6 addresses are 128 bits in length
  • Checksum
    1) Removed in IPv6
    2) This happens higher up in the header
  • Options
    1) Removed in IPv6
    2) IPv6 extension headers replace them
90
Q

Non-local IP routing

A
  • If the node cannot connect directly connect to its destination, this means that the destination is in a non-local network
  • As a result, the node will send the packet to a gateway which decides how to route the path of the data to the correct destination
91
Q

Local IP routing

A
  • If the destination node is on a network local to the host, then it can send packets directly
92
Q

IP Forwarding Algorithm

A
  • An implementation of IP routing
  • Uses a routing table to select the next destination for a datagram
  • If multiple destinations match, the route with the longest subnet mask is chosen
  • If multiple routes have the same subnet mask, the route with the lowest metric is used
  • Algorithm states given a destination IP address D and network prefix N:
    1) if(N matches a directly connected network address) {Deliver datagram to D over that network link;}
    2) else if(The routing table contains a route for N) {Send datagram to the next-hop address listed in the routing table;}
    3) else if(A default route exists) {Send datagram to the default route;}
    4) else{Send a forwarding error message to the originator;}
93
Q

Autonomous system

A
  • A collection of connected IP routing prefixes under the control of a single administrative entity/domain, each has a routing protocol to direct packets within (IGP)
94
Q

Distance-Vector Protocol

A
  • Each router computes distance between itself and each possible destination i.e. its immediate neighbors
  • The router share its knowledge about the whole network to its neighbors and accordingly updates table based on its neighbors
  • The sharing of information with the neighbors takes place at regular intervals
  • From this, it computes its own vectors of distances
  • Simple to use, but some protocols have a problem with slow convergence
95
Q

Link-State Protocol

A
  • Determines the best route for data packets based on cost
  • Each router shares knowledge of its neighbors with every other router in the network
  • A router sends its information about its neighbors to all the routers through flooding
  • Information sharing takes place only whenever there is a change
  • Using this information, the protocol computes maps
  • However, flooding could cause infinite looping which in turn, can be solved using TTL
96
Q

Slow Convergence

A
  • If there are changes to the network routing (for example, due to a broken link), then it requires many intermediate transfers of information to adjust to the route changes
97
Q

ARP

A
  • Stands for Address Resolution Protocol
  • Inhibit the sending of new TCP segments when new outgoing data arrives from the user if any previously transmitted data on the connection remains unacknowledged
  • ARP process:
    1) An Ethernet broadcasts an ARP request packet to all the machines on the LAN and asks if any of the machines know they are using that particular IP address
    2) When a machine recognizes the IP address as its own, it sends a reply so ARP can update the cache for future reference and proceed with the communication
    3) The machines that don’t recognise the IP address as their own simply discard the packet
98
Q

Gratuitous ARP

A
  • If a new machine joins the network or changes its IP it may send an ARP reply despite not being asked, all machines on the local network can read any ARP request and modify their own ARP caches
  • It is sent as a broadcast, as a way for a node to announce or update its IP to MAC mapping to the entire network
99
Q

DNS

A
  • Stands for Domain Name System
  • Main function is to translate domain names into IP Addresses, which computers can understand
  • They eliminate the need for humans to memorise IP addresses
  • Advantageous to use local DNS servers even forlhe local DNS server may have already looked up the domain name, therefore there’s no point wasting time in looking it up again
100
Q

Domain name format

A
  • The root of the tree is called . (dot) and this label is currently managed by the Internet Corporation for Assigned Names and Numbers (ICANN)
  • A Top Level Domain is a registry entity and its name is the name of the group to which its member-names are registered (for example, .com)
  • A Second Level Domain (directly under TLD) is commonly used to refer to the organisation that registered the domain name. It has a maximum letter length of 63 and can’t start or end with hyphens
  • A Subdomain is part of the main domain (for example, www)
  • Management is delegated to make
101
Q

Domain name management organisations

A
  • Labels under uk are managed by Nominet, the UK’s domain name registry
  • Labels under ac.uk are managed by Jisc, a digital service provider for UK-based educational institutions
  • Labels under bath.ac.uk are managed by University of Bath
  • Labels under cs.bath.ac.uk are managed by University of Bath’s Department of Computer Science
102
Q

Recursive Lookup

A
  • Running down the DNS tree sending SOA requests until the authority responsible for the host can return the IP or an error message
  • Sends request to local server
  • If nothing is returned, then send to a higher level
  • Example: for the link news.bbc.co.uk
    1) Look up who runs the uk label
    2) Look up who runs the co.uk label
    3) Look up who runs the bbc.co.uk label
    4) Look up who runs the news.bbc.co.uk label
103
Q

Reverse Lookup

A
  • Provides the domain name associated with a particular IP address
  • Configured by determining PTR records in a DNS server
  • The process involves searching domain name registry and registrar tables
  • A PTR record usually represents the IP entered backwards, followed by an in-addr.arpa entry
  • Example of use is tracking down a user sending spam emails
104
Q

Unreliable

A
  • Full transfer of data is not guaranteed
  • Unreliable protocols are appropriate for tasks where the speed of data delivery takes precedence over the accuracy of it, such as video streaming
105
Q

Connectionless

A
  • A signal goes out automatically without determing whether the receiver is ready, or even whether a receiver exists
  • In other words, there are no prior arrangements for sending data between endpoints
  • This type of protocol is considered unreliable
  • Example of CLP is UDP
106
Q

Connection-oriented

A
  • Data communication is established in which the endpoint devices use a set of preliminary protocols to establish end-to-end connections
  • This type of protocol is considered reliable
  • Example of COP is TCP
107
Q

Port

A
  • An endpoint of a communication
  • A 16-bit integer
108
Q

Ephemeral port

A
  • A temporary communication hub used for IP communications
  • It is created from a set range of port numbers by the IP software and used as an end client’s port assignment in direct communication with a well-known port used by a server
  • These are only valid for the duration of the communication session, and are free for reuse after the communication completes
109
Q

Well-known port

A
  • A port number that is reserved for common applications and privileged services (such as admin purposes)
  • Ports 0-1023 are assigned by Internet Assigned Numbers Authority (IANA)
  • Examples include:
    1) Port 80: HTTP server
    2) Port 21: File Transfer Protocol
    3) Port 25: SMTP
110
Q

Ethernet Frame

A
  • A term referring to an Ethernet data link header and trailer, plus the data encapsulated between the header and trailer
  • Has five fields
111
Q

Ethernet frame source address field

A
  • 6 bytes
  • Contains the MAC address of the machine where data is coming from
  • As Source Address is always an individual address (Unicast), the least significant bit of first byte is always 0
112
Q

Ethernet frame destination address field

A
  • 6 bytes
  • Contains the MAC address of the machine for which the data is travelling to
113
Q

Ethernet frame type field

A
  • 2 bytes
  • Indicates the type of the next layer up
114
Q

Ethernet frame data field

A
  • Between 46 and 1500 bytes
  • The place where actual data is inserted, also known as payload
  • Both IP header and data will be inserted here if IP is used over Ethernet
  • Needs to be at least 46 bytes since the frame must be long enough to detect a collision
  • If the data length is less than 46 bytes, then 0’s are added as padding to meet the minimum length
115
Q

Ethernet frame CRC field

A
  • 4 bytes
  • Contains a 32-bit hash code of data, which is generated over the other fields
  • If the checksum computed by destination is not the same as sent checksum value, data received is corrupted
116
Q

IPSec

A
  • Stands for Internet Protocol Security
  • Authenticates and encrypts the packets of data to provide secure encrypted communication between two computers over an IP network
  • Consists of three components: ESP (Encapsulating Security Payload), Authentication Header (AH) and Internet Key Exchange (IKE)
  • ESP provides data integrity through hash functions and confidentiality through encryption protection for IP packets, and only authenticates the payload
  • AH ensures connectionless integrity by using a hash function and a secret shared key in the AH algorithm, and authenticates the entire packet except for any mutable fields
  • IKE negotiates a key to deliver a secret key for the ESP and AH to work as well as sets up a security association (SA) in the IPsec protocol suite
  • Is good to use in that it allows TCP and UDP to be layered transparently on top of IPSec
117
Q

Problems with IPSec

A
  • There is an overhead at the first connection
  • Difficult to set up
  • Not as flexible as equivalent protocols such as TLS
118
Q

VPN

A
  • A private network that connects remote sites or users together over a public network
  • Allows a device to appear if it’s in a different location or network through tunnelling
  • Data encryption is added into the tunnel for the VPN to provide extra security as well as privatisation of a connection
  • Motivations for using a VPN include:
    1) Accessing internet content that is otherwise restricted due to location or censorship
    2) Businesses needing to access resources from outside its network
    3) To improve privacy by hiding the IP address from the ISP, since all activity would be routed through a VPN server
119
Q

Tunnelling

A
  • One protocol layers over another so the lower protocol can move the upper protocol transparently over a network that doesn’t typically carry the upper protocol
120
Q

Cons of VPN

A
  • A VPN prioritises security (i.e., hiding the IP address and encryption of data), meaning that data has to travel more which can have an effect on speed
  • Despite hiding internet activity from the ISP, this data is still accessible to the VPN provider
  • Some VPN providers have limits on data usage, therefore there ia a financial charge for increasing that limit
121
Q

TLS

A
  • Stands for Transport Layer Security
  • A stateful connection is established via a handshake to authenticate the connection and generate a session key
  • This session key is then used to encrypt any data being transferred through the connection
  • During the handshake, the server provides a digital certificate as a form of identification
  • The cerfiticate provides three details that allow authentication: the server name, some certificate authority and the server’s public encryption key
  • The session key is then generated and the handshake is concluded
  • Once this happens, the secured connection begins which is encrypted and decrypted with the session key until the connection closes
122
Q

Cons of TLS

A
  • There are several overheads:
    1) Encryption or decryption data creates overhead after every packet
    2) Setup messages and certificate validation creates overhead per connection
    3) Data expansion that occurs in the encryption process creates an overhead for every packet it happens in
  • The programmer needs to have strong understanding of how to set up the handshake connection
123
Q

Firewall

A
  • A security device that sits between a private network and the wider Internet, acting as a gateway
  • Decides whether to allow or block specific traffic based on a defined set of security rules
  • There are three common types of firewall: packet filters, application layer firewalls and application proxies
  • Most firewalls are software-based, but there is specialist hardware available
124
Q

Packet filter firewalling

A
  • Works in data link, network and transport layers
  • Inspects packets as they are transferred
  • Makes decisions on incoming and outgoing packets based on ports, protocols and the addresses of source and destination
  • One of three decisions are made on these packets:
    1) The firewall allows the packet to pass through
    2) The firewall silently discards the packet
    3) The firewall rejects the packet and generates an ICMP notification for the sender
125
Q

Application layer firewalling

A
  • Use a series of configured policies to determine whether to block or allow communications to or from an application
  • Typically works in the application layer
  • Different from traditional packet filters in that filters are applied by process rather than by port or protocol
  • Application firewalls can be either active or passive:
    1) Active firewalls actively inspect all incoming requests against known vulnerabilities such as SQL injections. Only requests deemed as “clean” are passed to the applications
    2) Passive firewalls inspect all incoming requests against known vulnerabilities, but they don’t actively reject or deny those requests if a potential attack is discovered
126
Q

Packet vs application layer firewalling

A

Pros of packet filtering:

1) They are fast, as they perform fewer evaluations
2) They are efficient, as just a single rule is able to help protect an entire network

Cons of packet filtering:

1) They are more susceptible to vulnerabilities, since they don’t inspect the payload of packets
2) They can be complex to configure

Pros of application filtering:

1) They are generally more secure, as they thoroughly inspect packet headers and data before distribution
2) They provide detailed logs of what the user is sending, allowing another way of monitoring network safety

Cons of application filtering:

1) Inbound data is processed by the application and by its proxy, as a result they are slower then picket filters
2) The logs produced by the firewall can be memory-intensive

127
Q

Privacy of a network connection

A
  • The protection and hiding of personal information from third parties
128
Q

Authentication of an network connection

A
  • The verification of a user that is requesting to connect to the network
  • Usually completed with a username and password
129
Q

Simple encoding

A
  • 0 voltage for 0 bit, 1 voltage for 1 bit
  • In principle, this is simple but there are problems:
    1) It is not possible to do carrier sense since a zero-stream looks exactly the same as an empty network
    2) Electrically it’s bad because a steady stream of 1 bits is basically a stream of 1V, which can destroy the wire the bits are travelling through
130
Q

Manchester encoding

A
  • Splits each bit period into two, and ensures that there is always a transition between the signal levels in the middle of each bit
  • The encoding of each data bit is either low then high, or high then low, for equal time
  • A 0 bit is represented by a low to high transition
  • A 1 bit is represented by a high to low transition
  • The main advantage of Manchester encoding is the fact that the signal synchronizes itself, this minimizes the error rate and optimizes reliability
  • The main disadvantage is the fact that this encoding doubles the frequency of the signal to 20 Mhz and can’t be used with 100Mb/s Ethernet rates
131
Q

Problems that physical encodings overcome

A
  • Prevents bits from drifting out of step
132
Q

MLT-3

A
  • Stands for Multi Level Transit
  • Like Manchester, transitions are used to encode bits
  • Cycles sequentially through 4 transitions: -1 to 0, 0 to 1, 1 to 0, 0 to -1
  • A transition encodes a 1 bit while a non-transition encodes a 0 bit
  • Used in conjunction with 4B5B, whose translations ensure that every set of 5 has a minimum of two transitions in order for the voltage to balance out to 0V
133
Q

4B5B

A
  • Groups 4 data bits into 5 physical bits for transmission
  • There is an issue of 25% more bits being required to send the same amount of information
  • Used in conjunction with MLT-3 for Ethernet to mitigate this
134
Q

ADSL

A
  • Stands for Asymmetric Digital Subscriber Line
  • Allows faster data transmission over copper telephone lines than a conventional voiceband modem can provide
  • Analogue telephone lines are capable of more than just voice communications but are limited to maximum speed of 56kb/s, and all frequencies are filtered out unless they fit into the range 300-3300Hz
  • Data transfer rate depends Increasing wire length decreases bandwidth
  • Is asymmetric in that available bandwidth is divided with a larger chunk for downloading and a small one for uploading
135
Q

First and last mile problem

A
  • The economic infeasibility of connecting every building to a telephone exchange using optical fibre
136
Q

UCS

A
  • Stands for Universal Coded Character Set
  • Uses 31 bits to represent a character
  • Contains over 136000 abstract characters, and each is identified through a unique name and a sequence of integer numbers called the code point
  • This allows the representation of many languages and scripts
137
Q

UTF-8

A
  • A character in UTF-8 can be from 1 to 4 bytes long and can represent any character in the Unicode standard
  • This allows ASCII values to be represented in UCS-compatible form since single bytes in the range 0-127 map directly to Unicode code points in the ASCII range
  • Is more compact than UTF-16 and doesn’t suffer from the same endianness issues as other UCS encodings
138
Q

Presentation Problem

A
  • Different encodings have different ways of interpreting bits
  • To ensure consistency on what these bits mean, they have to be translated so the original result is maintained
  • However, there is an issue of interpreting and changing bits that should not be changed
139
Q

Two Generals Problem

A
  • Two armies A and B, each led by a different general, are preparing to attack a city C
  • The armies are encamped near the city, each in its own valley
  • A third valley separates the two hills, and the only way for the two generals to communicate is by sending messengers through the valley
  • Unfortunately, the valley is occupied by C’s defenders and there’s a chance that any messages could be intercepted
  • A sends a message to B, stating the time they want to attack at dawn
  • It is required that the two generals have their armies attack the city at the same time in order to succeed, else A will fail the attack if they are alone
  • B has to send an ACK to A to confirm that they agree with the time
  • However, there is a possibility of the ACK being intercepted and A may not receive it
  • Because the ACK receipt can be lost as easily as the original message, a potentially infinite series of messages is required to come to consensus
140
Q

Retransmission timer

A
  • Used by TCP to avoid the Two Generals Problem
  • The sender starts a retransmission timer when sending a packet to some receiving host
  • The timer stops when the receiver sends an ACK back
  • If the timer expires before this, a timeout occurs, the timer is reset and the segment is retransmitted
  • If the network slows down (there is less bandwidth as a result of heavy traffic), then the RTT should increase
141
Q

RTT update rule

A
  • RTT = αRTT + (1 − α)M
  • M = actual round trip time
  • α = some smoothing factor
142
Q

SWS

A
  • Also known as silly window syndrome
  • Caused by poor TCP implementation
  • A serious problem can arise when the sending application program creates data slowly, the receiving application program consumes data slowly, or both
  • If a server with this problem is unable to process all incoming data, it requests that its clients reduce the amount of data they send at a time
  • If the server continues to be unable to process all incoming data, the window size shrinks and it can shrink to such an extent where the data being transmitted is smaller than the TCP Header
  • Since there is a certain amount of overhead associated with processing each packet, the increased number of packets means increased overhead to process a decreasing amount of data
143
Q

Solving SWS

A
  • Nagle’s Algorithm if SWS is created by the sender
  • Clark’s solution if SWS is created by the receiver
144
Q

Nagle’s algorithm

A
  • Inhibits the sending of new TCP segments when new outgoing data arrives from the user if any previously transmitted data on the connection remains unacknowledged
  • Improves the efficiency of TCP/IP networks by reducing the number of packets that need to be sent over the network
  • Works by combining a number of small outgoing messages (known as tinygrams) and sending them all at once in one large segment
  • As long as there is a sent packet for which the sender has received no acknowledgment, the sender should keep buffering its output until it has a full packet’s worth of output, thus allowing output to be sent all at once
  • Nagle and SWS fit together naturally, because when window scaling is in effect, “small” can’t be smaller than the window scale size
145
Q

Clark’s Solution

A
  • The receiver should not send a window update for 1 byte
  • The receiver should wait until it has a decent amount of space available or is half empty
  • The receiver should then advertise that window size to the sender
146
Q

Subvertable protocols

A
  • ARP, could pretend to be someone else therefore can be broken
  • DNS, fake websites (e.g. bank)

Year 3 (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions