Term
|
Definition
| A public network of networks which communicate through TCP/IP |
|
|
Term
| Why is the internet successful? |
|
Definition
1. Network Technology Independence - tcp/ip are independent of any particular brand or type of hardware. This allows the global internet to include a variety of netwrork technologies to communicate together.
2. Universal interconnection - tcp/ip allows any pair of computers to which it attaches to communicate. Each computer is assigned an address that is universally recognized throughout the internet. Every datagram carries the addresses of its source and destination. Intermediate switching devices use the destination address to make forwarding decisions.
3. End-to-end philosophy - tcp/ip internet protocols provides acknowledgements between the original source and ultimate destination instead of between successive machines along the path.
4. Application Protocol Standards - tcp/ip include standards for many common applications including electronic mail, file transfer, and remote login. Thus, when designing application programs that use tcp/ip, programmers often find that existing software provides the communication services they need. |
|
|
Term
|
Definition
| accomodates multiple, diverse underlying hardware technologies by providing a way to interconnect heterogeneous networks and a set of communication conventions that makes them interoperate. |
|
|
Term
| define "open systems interconnection" |
|
Definition
| open communication specifications that are publicly available, allowing anyone to build the software needed to communicate across an internet. |
|
|
Term
|
Definition
| syntactic and semantic rules for communication. They contain details such as message formats, computer responses to a message received, and error handling. |
|
|
Term
| define "application level internet "services |
|
Definition
| applications which run on top of tcp/ip. For example, WWW, email, file transfer, remote login |
|
|
Term
| define "etwork level internet services" |
|
Definition
basically tcp/ip, which provides two services:
1. connectionless packet delivery service - routing small messages from one computer to another based on address information carried in the message. does not guarantee in-order delivery because each packet is routed separately
2. reliable stream transport service - one computer establishes a connection with another to send large data. The underlying communication protocol divides the data into packets and sends them one at a time, waiting for receiver acknowledgement |
|
|
Term
| define "interoperability" |
|
Definition
| refers to the ability of diverse computing systems to cooperate in solving computational problems. |
|
|
Term
|
Definition
- data is broken into "packets"
- connectionless: no single path is reserved, each packet is sent by the best available path at the tim it is sent
- each packet contains the destination address |
|
|
Term
| define "circuit-switched" |
|
Definition
- data is sent as a continuous stream
- connection-oriented: a single path through the network is reserved for entire communication
- the network "knows" where the data is going |
|
|
Term
| what are the characteristerics of an internet technology? |
|
Definition
- addressing mechanism
- speed
- medium
- use of medium:
- encoding
- framing
- sending
- collision detection
- receiving
-error detection |
|
|
Term
| what are the characteristics of ethernet? |
|
Definition
- addressing mechanism: MAC address (6 bytes, 3 assigned to vendor and 3 assigned by vendor)
- speed: 10/100/1000Mbps
- medium: copper cables (cat-5, cat-6)
- use of medium:
- framing: 8-byte preamble of alternating 0's and 1's
- send: CSMA
- collision detection: CD |
|
|
Term
|
Definition
Carrier Sense Multiple Access/Collision Detection
- When a preamble is detected, check for bit errors, collisions, and for a match between the destination address and the receiving NIC
- if a full frame containing the receiving NIC arrives and if it has no bit errors, deliver the message to host |
|
|
Term
| what is the ethernet frame structure? |
|
Definition
48-byte frame:
- 8-byte preample
- 6-byte destination address
- 6-byte source address
- 2-byte frame type (used for interpreting frame. usually 01 for IP)
- 46-1500 bytes of data
- 4-byte CRC (Cyclic Redundancy Check) |
|
|
Term
| what is the ethernet exponential backoff ?policy |
|
Definition
| ethernet uses a binary exponential backoff policy where a sender delays a random time after the first collision, doubles the range if a second attempt to transmit also produces a collision, quadruples the range if a third attempt results in a collision, and so on. |
|
|
Term
| what is ethernet broadcast address? |
|
Definition
| all 1's address reserved for sending to all stations simultaneously |
|
|
Term
| what is the ethernet multicast address? |
|
Definition
provides a limited form of broadcast in which a subset of the computers on a network agree to listen to a given multicast address.
The set of participating computers is called a multicast group. To join a multicast group, a computer must instruct its host interface to accept the group's multicast address. |
|
|
Term
| what is network-level interconnection? |
|
Definition
| provides a mechanism that delivers small packets of data from their original source to their ultimate destination without using intermediate application programs. |
|
|
Term
| what are the advantages of network-level interconnection? |
|
Definition
- switching small units of data maps directly onto the underlying network hardware, making it extremely efficient
- separates data communication activities from application programs, permitting intermediate computers to handle network traffic without understanding the applications that are sending or receiving
- keeps the system flexible, making it possible to build general purpose communication facilities
-allows network managers to add new network technologies by modifing or adding a single piece of new network-level software, while application programs remain unchanged. |
|
|
Term
| how does information-hiding, abstraction, and separation of concerns apply to network-level interconnection? |
|
Definition
- separation of concerns: let gateways handle data communications and applications handle end-user concerns
- abstraction: packets are "object" and send and receive are operations of the application |
|
|
Term
|
Definition
- a unique address for each computer
- identifies the network and the computer in different parts of the address
- allows routing to a network rather than a host
- changeable, where nothing in the internet relies on any host having a permanent ip address |
|
|
Term
| define end-to-end argument |
|
Definition
- performance advantage: checking for correctness at ends is more efficient then checking at every link.
- correctness advantage: software errors cannot be detected from inside gateways |
|
|
Term
| what is dotted decimal notation? |
|
Definition
| four numbers separated by "." each with a range of 0 - 255. network id is first 3 numbers. last number is the host id. |
|
|
Term
| what is classful ip addressing? |
|
Definition
- breaks up the ip address space into blocks
- when an organization needs ip addresses, they get an entire block |
|
|
Term
|
Definition
- start with 0
-range 0-127
- net id is first octet
- host id is last 3 octets |
|
|
Term
|
Definition
- starts with 10
- range 128 - 191
- net id is first two octets
- host id is last two octets |
|
|
Term
|
Definition
- starts with 110
- range 192 - 223
- net id is first three octets
- host id is last octet |
|
|
Term
|
Definition
- multicast address
- starts with 1110 |
|
|
Term
|
Definition
- experimental
- starts with 1111 |
|
|
Term
| how many unique ip addresses can a single host have? |
|
Definition
| a unique address is assigned to each interface. so a host computer with several NIC's can have several unique addresses, with no upper bound limit. |
|
|
Term
| ip directed broadcast address |
|
Definition
- designated as all 1's in the host id
- packet is sent to all computers on a network |
|
|
Term
| ip limited broadcast address |
|
Definition
- all 1's address (32 1's)
- packet is sent to local network |
|
|
Term
|
Definition
- 32 0's for address
- used when a host needs to communicate but does not yet know its ip address |
|
|
Term
|
Definition
- where host id is all 0's
- example class B:
- Mask: 255.255.0.0
- network address: first two octets followed by all 0's
- example: 172.17.0.0 |
|
|
Term
|
Definition
- masks the network id of the ip address
- example class B:
- mask: 255.255.0.0 |
|
|
Term
|
Definition
|
|
Term
|
Definition
- should not be used for routing on the internet
- 10.0.0.0
- 172.16.0.0 ... 172.31.0.0
- 192.168.x.0 |
|
|
Term
|
Definition
- not enough addresses
- security
- address does not identify a computer |
|
|
Term
| accessibility issues with ip |
|
Definition
- choosing good alternate paths in case of failure can be time-consuming and hard
- using waypoints to shorten the paths that have to be found is not easy |
|
|
Term
|
Definition
- ipv4 uses big endian, meaning read from most significant byte first
-little endian reads from least significant byte first |
|
|
Term
|
Definition
| in networking, encapsulationis a technique in which a lower level protocol accepts a message from a higher level protocol and puts the message in its data portion. |
|
|
Term
|
Definition
the association of addresses at different layers
- for ip and physical layers: consider physical addresses as designating the NIC, then the problem is to "bind" and IP address to a NIC |
|
|
Term
| direct mapping of ip address to physical address |
|
Definition
if the size of a physical address on a network is small enough, then
- embed it in the ip address; or
- use a fixed function to map it to the ip address |
|
|
Term
| dynamic binding of ip address to physical |
|
Definition
| assign physical and ip addresses independently, and use a protocol to determine mapping of ip address to physical address |
|
|
Term
| characteristics of a protocol |
|
Definition
- formal name
- problem(s) solved
- general idea
- message formate(s)
- sequence of actions and messages |
|
|
Term
|
Definition
| Address Resolution Protocol |
|
|
Term
|
Definition
| translating an ip address to a physical address; allowing ip addresses to be arbitrarily assigned |
|
|
Term
|
Definition
- to send a message to an ip address:
- broadcast a request message: "Who has ip address x.x.x.x?"
- the computer having that address replies: "I have IP address x.x.x.x and my physical address is pppppp."
- store the result in the ARP cache (a table of IP vs physical addresses |
|
|
