Principle of three-layer switching technology

Simply put, the three-layer switching technology is: two-layer switching technology + three-layer forwarding technology. It solves the problem of network bottleneck caused by the low speed and complexity of the traditional routers after the network segment is divided and the network segment has to rely on the router for management. 1, what is the three-tier exchange

The three-tier exchange (also known as multi-layer switching technology, or IP switching technology) is presented in relation to the traditional exchange concepts. As we all know, the traditional exchange technology is the second layer in the OSI Network Standard Model-the data link layer operates, and the three-layer switching technology is the third layer in the network model to implement the high-speed packet forwarding. Simply put, the three-layer switching technology is: two-layer switching technology + three-layer forwarding technology.

The emergence of the three-layer switching technology solves the problem of network bottleneck caused by the low speed and complexity of the traditional routers, after solving the division of the network segment in the LAN, the network segment must rely on the router for management. 2, three layer Exchange principle

A three-layer switching device is a second-tier switch with a third-layer routing function, but it is an organic combination of the two, not simply to overlay the hardware and software of the router device on the LAN switch.

The data communication process is:

Site A and B communicate through the three-tier switch. Sites A and B have segments that belong to the direct network segment on the switch, if site A and site B are not in the same subnet, send station a first to its "default gateway" issue ARP Request message, and "Default gateway" IP address is actually the three-tier switch on the IP address of the VLAN owned by site A.

When send station A to the "Default gateway" IP address broadcast an ARP request, the switch to send station a back to an ARP reply message, told the site a switch this VLAN MAC address, at the same time through the software to the site a IP address, MAC address, Information such as the port number directly connected to the switch is set to the three-tier hardware table entry for the switching chip.

When site a receives this ARP reply message, the destination MAC address is replaced and the packet to be sent to B is first sent to the switch. After the switch received this package, the same first to the source MAC address learning, the purpose of MAC address lookup, because the purpose of the MAC address is the switch MAC address, in this case will send the message to the Exchange chip three-tier engine processing.

In general, the three-tier engine will have two tables, one is the host routing table, which is indexed by IP address, which holds the destination IP address, the next hop MAC address, port number and other information. If a matching table entry is found, the message is forwarded from the specified port in the table after some manipulation of the message (for example, the destination Mac and the source Mac replacement, TTL minus 1, and so on).

If a matching entry is not found in the host routing table, it continues to look for another table-the Network Segment routing table. This table holds the network segment address, next hop MAC address, port number and so on information. Generally speaking, this table is a lot less entries, but the scope of coverage is very large, as long as the set properly, basically can ensure that most of the packets into the switch hardware forwarding, so that not only greatly improve the forwarding speed, but also reduce the CPU load.

Since the three-tier engine inside the chip has already saved site A, B of the routing information, the site A, b between the communication or other segments of the site want to communicate with a, B, switching chips will be directly from the three-tier hardware table items specified in the port forward, and do not have to give the packet to the CPU processing. This through "One route, multiple Exchange" way, greatly improve the forwarding speed.

Three-layer Exchange from the concept of the proposed to today's popular applications, although only after a few years of time, but its application in the network construction is more and more extensive, from the initial backbone layer, the middle of the convergence layer has been penetrated to the edge of the access layer. The three-layer switch has many advantages, such as fast speed, good performance and low price.

Where there is no WAN connectivity requirements, but also the need for routers, can be replaced by a three-tier switch. With the development of ASIC hardware chip technology and the popularization of practical application, the three-layer exchange technology and products will be further developed. 3, Cisco three-layer Exchange product characteristics:

In general, the third tier exchange products are programmable scalable ASIC chip technology, can provide some of the following rich features:

(1) Non-blocking line-speed switching and routing for all network interfaces and protocols on all ports;

(2) with very high throughput, packet forwarding speed (that is, the transfer/second, PPS) is usually faster than the high-end router 10~100 times;

(3) A variety of protocol routing options, such as IP (RIPV1/V2, OSPF), IP multicast (DVMRP, PIM) and IPX;

(4) Support a variety of VLAN division, according to the Port/mac address, protocol, IP subnet, IEEE 802.1Q or Cisco ISL Division;

(5) Bandwidth reservation (RSVP) with service class (CoS) and quality of service (QoS) Priority processing, support IEEE 802.1p and Business Classification (DIFFERSERV);

(6) Access list control can be set filter rules, or firewall based security policy;

(7) Support Point-to-Point Protocol (PPPoE) through Ethernet, support security user authentication, cooperate with user billing, enhance user management characteristics;

(8) To support the increase of bandwidth allocation service of Ethernet;

(9) ASIC programmability, supporting technologies such as IPV6 and other future technologies to protect user investment. 4. Advantages

In the past, most of the data in the network complied with the "80/20" rule, that is, only about 20% of the data packets in the network communicate with the central server or other parts of the enterprise network through the backbone routers, while 80% of the network traffic is still concentrated in different department subnets. Now, however, the situation has changed radically, creating a "20/80" rule. In order to cope with the increasing data flow, the network of shared media is replaced by switched networks. This change has a direct impact on traditional routers used in network segmentation. Since most of the data traffic spans IP subnets, routers have actually become the bottleneck of network transmission.

The traditional router's main function is to realize routing and network interconnection, that is to obtain the topological information of the subnet and the network characteristic of each physical line through certain way, and obtain the best path of each subnet through a certain routing algorithm, establish the corresponding routing table, so that each IP packet jumps to jump (hop to hop) and reaches the destination; Second, it must deal with different link protocols. When IP packets pass through each router, they need to be processed by the software such as queuing, protocol processing and addressing selection routing, which causes the delay to increase. At the same time, the router adopts shared bus mode, the total throughput is limited, and when the number of users increases, each user's access rate decreases. Routers are more focused on support for multiple media types and multiple transmission speeds, and the current data buffering and switching capabilities are more important than line-speed throughput and low latency. Although the performance of the router has recently been improved, about 1Mpps, but the cost of using this router is surprisingly high.

Compared with routing technology, the advantage of switching technology is fast, when the network scale is very large, high-speed, high-capacity routers is very necessary. On the other hand, because the modern communication network mostly uses the optical fiber technology, therefore the data network main bottleneck is the node router now. The current third-tier exchange, routing exchange, or other nouns are the result of this mentality. Although the third-tier exchange was originally designed for the LAN, it uses the destination IP address for Exchange, but now this technology has been used in the WAN.

Third-tier exchange plays a more and more important role in the current network construction, it does not need to broadcast packet proliferation, but directly using the dynamic establishment of the MAC address to communicate, such as IP address, ARP and so on, with multicast and virtual network based on IP and IPX protocol routing functions, this aspect of the function of the smooth realization, Mainly rely on specialized integrated circuits (ASIC). The traditional routing software is changed to the embedded instruction of ASIC, which accelerates the forwarding and filtering of packets, so that the linear Routing and service quality are guaranteed reliably at high speed.