Basic principles of layer 2 switching

Basic principles of layer 2 switching

Learn from the Ethernet layer 2 source through the Ethernet layer 2 switch MAC To maintain the MAC The correspondence between address and port （ preservation MAC The table corresponding to the port is called MAC surface ）, Through its purpose MAC Search for MAC Vote on which port to forward , The basic process is as follows ：

（1） The layer 2 switch receives the Ethernet frame , The source MAC Write the corresponding relationship with the receiving port MAC surface , As the basis for layer 2 Forwarding in the future . If MAC The same entry already exists in the table , Then refresh the aging time of the table item .MAC Certain aging and updating mechanism shall be adopted for table items , Items that have not been refreshed within the aging time will be deleted ;

（2） According to the purpose of Ethernet frame MAC Go find MAC surface , If no matching table entry is found , Then forward to all ports （ Except for receiving port ）; If the purpose MAC It's a broadcast address , Then forward to all ports （ Except for receiving port ）; If you can find a matching table entry , Forward to the corresponding port shown in the table entry , However, if the port shown in the table item is the same as the port receiving the Ethernet frame , Then discard the frame .

As can be seen from the above process , Layer 2 switching is maintained through MAC Table and according to the purpose MAC Table lookup forwarding , Effective use of network bandwidth , Improved network performance .

chart 4 Is an example of layer 2 switching

                          chart 4 Example of layer 2 switching

Layer 2 switch MAC Address aging and refreshing are usually done directly by the hardware ASIC Chip to complete , General software implementation of network bridge . By the way, here is a brief introduction to the mechanism it usually adopts ：

（1） Every dynamically added... Stored in the chip MAC Each address table entry has a 1 bit Aging mark of length , At the same time, the chip has an aging timer to control address aging ;

（2） For the newly learned MAC Address table entry , Its aging mark position 1; For what has been learned MAC Table item , If there is a message source in the follow-up MAC Same as table item , Then refresh its aging flag bit to 1;

（3） Whenever the chip aging timer times out , take MAC The aging flag bit in the address table is equal to 1 Project , Modify its aging flag bit to be equal to 0; about MAC The aging flag bit in the address table is equal to 0 Project , Delete directly .

Under such an aging and updating mechanism ,MAC The actual aging time of the address is not accurate , It's a range ：1～2 Times the aging timer time .

Although the layer-2 switch can isolate the conflict domain , But it can not effectively divide the broadcast domain . As can be seen from the forwarding process of the layer-2 switch described above , Broadcast message and purpose MAC If the search fails, the message will be forwarded to all ports , When the number of hosts in the network increases , This situation will consume a lot of network bandwidth , And it also brings a series of problems in terms of security . Of course , One way to isolate broadcast domains is through routers , However, due to the high cost and low forwarding performance of routers, the application of this method is limited . Based on these situations , In layer 2 switching VLAN technology .
————————————————
Copyright notice ： This paper is about CSDN Blogger 「SalaWang」 The original article of , follow CC 4.0 BY-SA Copyright agreement , For reprint, please attach the original source link and this statement .
Link to the original text ：https://blog.csdn.net/erwangshi/article/details/45038837

Ethernet layer 2 switching foundation
https://www.doc88.com/p-5791211246520.html?r=1
https://blog.csdn.net/erwangshi/article/details/45038837
https://zhuanlan.zhihu.com/p/339326516