Reading notes-"Graphic TCP/IP" (2/4)

Classic Excerpt chapter III Data link

1, the actual communication medium between the processing is the high and low voltage, the flash of light and the strength of the electric wave signal.
2. Data link layer Related technologies: Mac addressing, media sharing, non-public network, packet switching, loop detection, VLAN, etc.
3. The data link layer is considered as the smallest unit in the network transmission.
4, Ethernet, FDDI and ATM differences
Ethernet and FDDI not only contain the second layer of data link layer of the OSI Reference Model, but also specify the specifications of the first layer of physical layer.
The children in the ATM include part of the third layer of the network layer.
5. Network topology: The connection and form of the network.
6, the MAC address is 48 bits long.

7, from the use of communication media, the network can be divided into shared media type and non-shared media type. A shared media network refers to a network where multiple devices share a single communication medium. The earliest Ethernet and FDDI are media-shared networks with half-duplex communication. There are two types of media access control in a shared media network: Contention mode and token delivery mode. Contention means competing for the right to acquire data, called CSMA (carrier-monitored multi-access). Token delivery is a special message called a "token" along the token ring, which is a way of controlling transmission. Only the station that obtains the token can send data. The features are: (1) There will be no conflict (2) each station has the opportunity to obtain tokens through an equal cycle.

8. Token Append: The token is sent to the next station without waiting for the receiver's data to arrive to confirm.
9, non-shared media network: The sending side and the receiving end do not share the communication media, the use of full-duplex communication mode.
10. The switching hub is also known as an Ethernet switch. An Ethernet switch is a bridge that holds multiple ports.
11, switch two kinds of forwarding: storage and forwarding-Avoid sending error frames, pass-through forwarding-the delay is short, but inevitably there is the possibility of sending the wrong frame.
12, loop detection technology in two ways: (1) Spanning Tree mode (2) source routing mode. Spanning Tree mode: Each bridge must exchange BPDU packets every 1-10 seconds to determine which ports are using which ports are not used in order to eliminate loops. Source Routing method: To determine the source address of the sending data is transmitted through which bridge, the sending side itself must have the function of the source route.
13. If the transmission speed is the same and the cable used is different, you can connect the repeater or hub that allows the replacement of the transmission medium. In the case of different transmission speeds, you must use devices such as bridges, switch hubs, or routers that allow change speed.
14. The preamble indicates the beginning of an Ethernet frame, and it is the flag that the peer network card can be synchronized with. At the end of the preamble is a field called SFD, whose value is "11". The maximum data range that a data frame can hold is 46~1500 bytes. The end of the frame is a 4 byte called FCS.
15, the Data link layer is divided into the media access control layer and the logical link control layer.
16, IEEE802.11 in the physical layer using electromagnetic or infrared.
17, WiMAX is the use of microwaves in the enterprise or the home to achieve wireless communication of a way. ZigBee is mainly used for remote control of home appliances, which is a short-distance, low-power wireless communication technology.
18. The main functions of PPP include two protocols:

(1) does not rely on the upper-layer LCP protocol: mainly responsible for establishing and disconnecting, setting the maximum receiving unit, setting the authentication protocol and setting whether to monitor the quality of communication.
(2) dependent on the upper-level NCP protocol (also known as IPCP): Responsible for IP address settings and whether TCP/IP header compression and other settings.

19, ATM is a unit called cell transmission of the data link, due to its short line occupancy time and the ability to efficiently transfer large-capacity data and other characteristics are mainly used for wide-area network connection.
20, Aym so far the biggest drawback: even if only one cell error should be discarded all packets, to re-send.
21. Pos is a kind of protocol for Packet communication on SDH of synchronous digital system.
22, FDDI is called Distributed optical fiber data interface. Using Token Ring access mode, Token Ring access mode in the case of network congestion is very easy to lead to network convergence.
23, HDMI indicates high-definition multimedia interface. DOCSIS is the industry standard for cable TV transmission data.

24, optical fiber to the FTTH, fiber to the building FTTB.
25. VPN Virtual private network is used to connect the remote area. Includes Ip-vpn and wide-area Ethernet.

Fourth Chapter

1, the definition of the host: configured with an IP address, but not routing control of the device.
Router: A device that has both an IP address and a routing control capability.
Nodes are collectively referred to as hosts and routers.
2. Routing control: the ability to send packet data to the final destination address.
3, a jump refers to the use of the data link layer below the hierarchical function of the transmission of a data frame interval.
4. The biggest difference between different data links is that they have their own maximum transmission units.
5, IP for no connection, that is, self-love before the contract, do not need to establish a connection to the target address. Why is IP oriented to no connection? (1) In order to simplify (2) in order to speed up.
6, the classification of IP address:

(1) Class A address: the first address that begins with "0", 0.0.0.0~127.0.0.0 is the network address of Class A.
(2) Class B address: First address with "10", 128.0.0.0~191.255.0.0 is the network address of Class B.
(3) Class C address: The first address that begins with "110", 192.0.0.0~239.255.255.0 is the network address of Class C.
(4) Class D address: First address starting with "1110", 224.0.0.0~239.255.255.255 is the network address of Class D.

7. Broadcasting is divided into local broadcasting and direct broadcasting. Local broadcasts are broadcasts within the network, and direct broadcasts are broadcast between different networks.
8, non-type domain routing CIDR: The network identity and the host identity of the IP address are separated by any length.
9. Private IP:
10.0.0.0~10.255.255.255 (10/8) Class A
172.16.0.0~172.31.255.255 (172.16/12) Class B
192.168.0.0~192.168.255.255 (192.168/16) Class C
In addition to the above private IP addresses are global addresses.
10, the Routing Control Table Formation mode: (1) The administrator manually set (2) the router and other routers exchange information with each other automatically refresh.
11. The default route refers to a record in which any address in the routing table can match. Generally marked as 0.0.0.0/0.
12, loop Address: 127.0.0.1. The same value as this address is a host name called LocalHost. When this IP or hostname is used, the packet does not flow to the network.
13. Path MTU Discovery: Refers to the maximum MTU size when shards are not required from the sending side host to the receiving end host.
14, the characteristics of IPV6:

(1) Expansion of IP address and aggregation of routing control tables
(2) Performance improvement: simplifies the header structure and reduces the load on the router
(3) Support Plug and Play function, automatically assign IP address
(4) Using authentication and encryption function
(5) Multicast, Mobile IP becomes the extension function
15. The only local address is the address used when the Internet communication is not performed.

Feel

This week I mainly read the third fourth chapter, the content of these two chapters is to explain the data link layer and the network layer of the Protocol, and the composition of each protocol packet structure has a detailed explanation, let me from the microscopic perspective of the two layers of the function of a comprehensive understanding. By looking at the contents of these two chapters, I think my biggest gain is figuring out that the data Link layer protocol is used for communication between two nodes in the network layer routing addressing process. This is the place where I feel enlightened. Also is to have and before the same understanding, the book read times its righteousness from the present, the book read hundreds taught. It was a bit exaggerated, but it did get something.

Reading notes-"Graphic TCP/IP" (2/4)