Computer Network review

1. networks refer to the "three networks", that is, telecommunications networks, cable TV networks, and computer networks.
2. A network is composed of several nodes and links connecting these nodes.
3. People regard 1983 as the birth time of the Internet.
4. Differences between Internet and Internet
• The internet (internet or internet), which starts with the lowercase letter "I", is a general term. It refers to a network formed by the interconnection of multiple computer networks.
• Internet, which starts with an uppercase letter (I), is a special term. It refers to the world's largest, open, and interconnected computer network, it uses the TCP/IP protocol family as the communication rule, and its predecessor is the ARPANET of the United States.
5. Three-Level Computer Networks are divided into backbone networks, regional networks, and campus networks (or enterprise networks ).
The third stage is characterized by the gradual formation of a multi-layer ISP-structured Internet
6. Composition of the Internet
(1) The edge part consists of all hosts connected to the Internet. This part is directly used by users for Communication (Transfer Data, audio or video) and resource sharing.
(2) The core part consists of a large number of networks and routers connected to these networks. This part provides services for the edge part (providing connectivity and exchange ).
7. Two communication modes
• Client/Server (C/S)
• Peer-to-Peer (P2P)
8. A vro is a key component for packet switching. Its task is to forward and receive packets. This is the most important function in the core part of the network.
9. Three Phases of circuit switching: establishing connections, communicating, and releasing connections
10. The vro processing group process is as follows:
-Put the received group into the cache (temporarily stored );
-Query the forwarding table to find the port from which the destination address should be forwarded;
Send the group to the appropriate port for forwarding
11. Advantages of group exchange:
• Efficiently and dynamically allocates transmission bandwidth, which occupies segments of communication links.
• Flexible group-based transmission units and search routes.
• Quickly send groups to other hosts without establishing a connection.
• Reliable and reliable network protocols; distributed routing selection protocols enable the network to survive.
12 problems caused by group exchange
• Groups need to be queued for storage and forwarding at each node, which may cause a certain latency.
The header that a group must carry (containing necessary control information) also causes certain overhead.
13. several different types of networks:
-WAN (Wide Area Network)
-LAN (Local Area Network)
-MAN (Metropolitan Area Network)
-Personal Area Network (PAN)
Networks of different users:
-Public network)
-Private network)

14. Speed
• Bit is the unit of data volume in a computer and the unit of information used in information theory.
• Bit is derived from binary digit, meaning a "binary number". Therefore, a Bit is one or zero of binary numbers.
• Data rate or bit rate is one of the most important performance indicators in computer networks. The unit of rate is B/s, or kb/s, Mb/s, Gb/s, etc.
• Rate usually refers to the rated rate or nominal rate.
15. bandwidth
• "Bandwidth" originally refers to the frequency band width of a signal, in the unit of HEH (or qihe,, gehe, etc ).
"Bandwidth" is the equivalent of the "highest data rate" that digital channels can transmit. The unit is "bit per second" or B/s (bit/s)
Total latency = Sending latency + propagation latency + Processing latency + queuing latency
16. Architecture of layer-5 protocols
• Application layer)
• Transport layer)
• Network layer)
• Data link layer)
• Physical layer)
Chapter 2 Physical Layer
1. parallel transmission of data in computers
2. Features of the Physical Layer
• Mechanical characteristics indicate the shape and size of the connectors used for the interface, number and arrangement of leads, and the fixing and locking devices.
• Electrical characteristics indicate the range of voltages that may occur on each line of the interface cable.
• Function features indicate the meaning of the voltage of an electric flat on a line.
• Process features indicate the sequence in which various events of different functions appear.
Channel Multiplexing Technology
1. frequency division multiplexing, time division multiplexing and statistical time division multiplexing, wavelength division multiplexing, and code division multiplexing
Chapter 3 data link layer
The data link layer uses the following channels:
Point-to-point channel
One-to-one point-to-point communication
Broadcast Channel
One-to-many Broadcast Communication
In addition to physical lines, the data link must also have a communication protocol to control the transmission of the data. If the hardware and software implementing these protocols are added to the link, a data link (logical) is formed ).
The hardware and software used by the ENI to implement these protocols. Generally, the NIC includes the data link layer and the physical layer.
The Data Exchange Unit of the physical layer is binary bit.
Protocol Data Unit-Frame
Protocol data units in the listen network layer are called IP Datagram
The main steps of the  point-to-point data link layer communication are as follows:
The data link layer of node A encapsulates the IP data packets handed over by the network layer into frames at the header and end;
Node A sends the frame to Node B;
If no frame error is received at the data link layer of Node B, IP datagram is extracted from the frame and handed over to the network layer; otherwise, the frame is discarded.
• Basic Problems to be Solved during data transmission at the data link layer
-Encapsulated into frames
-Frame Synchronization (transparent transmission, using the byte filling method)
-Error Control (error code check and Error Correction Code)
Maximum transfer unit (MTU)

-Error Correction Code (check error code)
• CRC Cyclic Redundancy Test
• (N, k) Code
• Concise and practical exclusive or (XOR)
-Error types and response policies during transmission
• Frame loss, frame duplication, and frame out-of-order
• Frame number, confirmation, and retransmission
Currently, the most widely used data link layer Protocol in the world is Point-to-Point (PPP)
• Requirements to be met by the PPP protocol
-Simplicity
-Encapsulated into frames
-Transparency
-Multiple network layer protocols
-Multiple types of links
-Error Detection
-Check the connection status.
-Maximum transfer unit
-Network layer address negotiation
-Data Compression negotiation
• Transparent transmission of PPP protocols
-Asynchronous transmission
• Convert each 0x7E in the information field into a 2-byte sequence (0x7D, 0x5E ).
• If a 0x7D byte appears in the information field, convert it into a 2-byte sequence (0x7D, 0x5D ).
• If the ASCII control character (that is, the value is less than 0x20) is displayed in the information field, a 0x7D byte must be added before the character, at the same time, the encoding of this character is changed.
-Synchronous transmission
• If synchronous transmission is used (a series of BITs are continuously transmitted ). In this case, the PPP protocol uses the zero-bit padding method for transparent transmission.
On the sender side, if five consecutive values are found to be 1, enter 0 immediately. The receiver scans the bitstream in the frame. Each time five consecutive values are found to be 1, the value 0 after these five consecutive values is deleted.

• Media Sharing Technology
-Static channel division
• Frequency Division Multiplexing
• Time division multiplexing
• Wavelength Division Multiplexing
• Code Division Multiplexing
-Dynamic Media Access Control (Multi-Point Access)
• Random Access
Controlled access, such as polling or polling
-In order to make the data link layer better adapt to various LAN standards, the 802 committee split the data link layer of the LAN into two child layers:
-Logical Link Control (LLC) Sub-Layer
-Medium Access Control (MAC) Sub-layer.
-Content related to access to the transmission media is placed on the MAC Sub-layer, while the LLC sub-layer is irrelevant to the transmission media. No matter what protocol is used, the LAN is transparent to the LLC sub-layer.
Functions of the adapter
-String and conversion
-Cache data
-Ethernet protocol implementation
Therefore, the coordination method adopted by Ethernet is a special protocol-CSMA/CD (carrier listening Multi-Point Access/collision detection)
-"Multi-Point Access" indicates that many computers are connected to a bus through multi-point access.
-"Carrier listener" means that each station must first check whether other computers on the bus are sending data before sending data. If so, do not send data temporarily to avoid collision.
-"Collision detection" refers to the signal voltage of the computer while sending data while detecting the channel.

• Competition period
-Record one-way end-to-end propagation latency on the bus as t
-The station that sends the data frame first knows whether the sent data frame is subject to a collision after a maximum of 2 TB (twice the end-to-end round-trip latency) after the data frame is sent.
-The end-to-end round-trip latency of Ethernet is 2 TB, which is called a contention period or a collision window.
After the competition period, the collision has not been detected, so that it is certain that this transmission will not have a collision.
Length of the competition period
-Ethernet takes 51.2 MS as the length of the contention period.
-For 10 Mb/s Ethernet, 512 bits (64 bytes) can be sent during the contention period.
-When data is sent over Ethernet, if the first 64 bytes do not conflict, the subsequent data will not conflict.
Binary Index type Backoff Algorithm
-After a collision station stops sending data, it must delay (backoff) a random time before sending data again.
• Determine the basic backoff time. Generally, it is set to a competition period of 2 TB.
• Defines the number of retransmission times k and £10, that is
K = Min [retransmission times, 10]
• From the Integer Set [0, 1 ,..., (2 k-1)], a random number is taken as r. The latency required for retransmission is the basic backoff time of r times.
• The frame is discarded and reported to the High level when the frame is successfully re-transmitted 16 times.
Shortest valid frame length
• If a conflict occurs, it must be within the first 64 bytes of the message.
• As soon as a conflict is detected, the sending is aborted immediately, and the data already sent must be smaller than 64 bytes.
• Ethernet specifies the minimum valid frame length of 64 bytes. Any frame whose length is smaller than 64 bytes is an invalid frame aborted due to a conflict.
Enhanced collision
When the site that sends data finds a collision:
Stop sending data immediately;
Continue to send several bits of human interference signals (jamming signal) so that all users know that a collision has occurred.
A hub is essentially a multi-interface forwarder. It simply forwards bits and works on the physical layer without collision detection.

Invalid MAC frame
-The frame length is not an integer byte;
-An error was found using the received frame inspection sequence FCS;
-The length of the data field is not 46 ~ 1500 bytes.
For MAC frames that are invalid, the system simply discards them. Ethernet is not responsible for re-transmitting discarded Frames
• Extended Ethernet at the data link layer
-The LAN extension on the data link layer uses bridges.
-The bridge works on the data link layer. It forwards received Frames Based on the target address of the MAC frame.
-The bridge can filter frames. When the bridge receives a frame, it does not forward the frame to all interfaces. Instead, it checks the target MAC address of the frame and then determines the interface to which the frame will be forwarded.
• Advantages of Bridges
-Filter traffic.
-Expanded the physical scope.
-Improved reliability.
Lan that can interconnect different physical layers, different MAC Sub-layers, and different rates (such as 10 Mb/s and 100 Mb/s Ethernet)
Disadvantages of Bridges
-Storage forwarding increases latency.
-There is no traffic control function on the MAC sublayer.
-A bridge is only suitable for a LAN with fewer users (no more than a few hundred) and less traffic. Otherwise, network congestion may occur due to the spread of too much broadcast information. This is the so-called broadcast storm.
High-speed Ethernet
100BASE-T Ethernet
100 represents a maximum rate of 100 MB/S, and T represents a twisted pair.
1000BASE-X/SX/LX/Cx Ethernet
X represents optical fiber, SX represents short-wave length, LX long wavelength, CX copper wire.
Chapter 4: Network Layer

 

• For more effective search, a route table without a classifier address is usually stored in a hierarchical data structure and then searched from top to bottom. The most common here is binary trie ).
Internet Control Packet protocol ICMP
• PING is used to test the connectivity between two hosts.
• PING uses ICMP send-back request and send-back response packets.
• PING is an example of using network layer ICMP directly at the application layer. It does not use TCP or UDP at the transport layer.
• The internal Gateway Protocol IGP (Interior Gateway Protocol) is the routing Protocol used within an autonomous system. (RIP, OSPF)
• External Gateway Protocol EGP (External Gateway Protocol ). The routing protocol used between autonomous systems. BGP-4)
• Working Principle
• RIP is a distributed routing selection protocol based on distance vectors.
• The RIP Protocol requires that every vro in the network maintain a distance record from itself to every other destination network.
• The distance from a vro to a directly connected network is defined as 1.
• The distance from a router to a non-directly connected network is defined as the number of routers plus 1.
• The distance in the RIP Protocol is also known as hop count, because each hop passing through a vro adds 1.
• The distance actually refers to the "shortest distance ",
• RIP considers a good route to have a small number of routers, that is, a short distance ".
• RIP allows a maximum of 15 routers.
• When the maximum value of "distance" is 16, it is equivalent to inaccessible. It can be seen that RIP is only applicable to small Internet.
• RIP cannot use multiple routes between two networks at the same time. RIP selects a route with the least router (that is, the shortest route), even if there is another route with high speed (low latency) but many routers.
• Only exchange information with adjacent routers.
• The information exchanged is all information that the current vro knows, that is, its route table.
• Exchange route information at a fixed interval, for example, every 30 seconds.
•

• The biggest advantage of the RIP Protocol is its simplicity and low overhead.
• RIP limits the size of the network. The maximum distance it can use is 15 (16 indicates inaccessible ).
• The route information exchanged between routers is the complete route table in the router. As the network grows, the overhead increases.
• One problem with RIP is that when a network failure occurs, it takes a long time to transmit this information to all routers.
OSPF uses reliable flood methods.
Therefore, when the Internet is large, the OSPF protocol is much better than the vector protocol RIP.
External Gateway Protocol BGP
BGP is a protocol for exchanging routing information between vrouters of different Autonomous Systems.
The Border Gateway Protocol (BGP) can only be used to find a route that can reach the destination network and is better (not in circles), rather than finding the best route.
The node order of the BGP protocol to exchange route information is the magnitude of the number of autonomous systems,
A vro is a dedicated computer with multiple input and output ports. Its task is to forward packets.
Multicast use group address -- IP uses Class D address to support multicast. Multicast addresses can only be used for the destination address, but not for the source address.
Virtual Private Network VPN and Network Address Translation NAT Virtual Private Network VPN
Chapter 5
The transport layer provides communication services to the application layer above it. It belongs to the highest level of the communication-oriented part and is also used
The lowest layer in the User Function.
The transport layer must have two different transport protocols: connection-oriented TCP and connectionless UDP.
When the transport layer adopts the connection-oriented TCP protocol, although the following network is not reliable (only provide the best service ), however, this logical communication channel is equivalent to a full-duplex reliable channel.
When the transport layer uses a connectionless UDP protocol, this logical communication channel is an unreliable channel.
• The data unit protocol transmitted over TCP is a TCP segment (segment)
• The data unit protocol transmitted by UDP is UDP packets or user datagram.
The abstract protocol port in the protocol stack layer is the software port.
The port on the vro or vswitch is a hardware port.
UDP only adds a few features to the IP datagram service, namely the port function and error detection function.

Author: "ITeamsky-Yang Bo's technical space"