Shulou(Shulou.com)06/01 Report--

According to the connection mode, WAN connections can be divided into the following three types:

According to the exchange mode of data units, it can be divided into the following three types:

Circuit switching

The main feature of circuit switching is that it requires the establishment of an actual physical path between the two sides of the communication, and this path is monopolized in the whole communication process. The most common example of circuit switching is a telephone system, such as PSTN (Public Service telephone Network).

Message exchange

A kind of storage exchange, the so-called "storage exchange" means that before data exchange, it is cached through the buffer memory and then processed according to the queue. "storage exchange" is divided into "message switching" (Message Switching) and "packet switching" (Packet Switching).

The basic idea of message switching is to store the user's message in the memory of the switch, and then send the message to the receiving switch or user terminal when the required output circuit is idle.

Packet switching

This method is similar to message exchange, but the message is sent in groups, and the maximum length of the packet is specified. After arriving at the destination, the packet needs to be reassembled into a message. Both X.25 and frame Relay use this switching technology. Now let's get to the topic of today.

Frame Relay technology is a relatively old WAN connection method, and it is also a WAN encapsulation protocol, which is seldom used by enterprises at present. Of course, many bank ATM machines are still using this service (it meets the business requirements of security, stability, reliability and scalability, low transmission rate and low cost).

Frame Relay technology is a technology that uses a simplified method to transmit and exchange data units on the second layer of OSI, using the form of frames to encapsulate users' data for cross-network transmission. The frame Relay network is generally built by the public information service provider (ISP). For customers who need this service, they can go to the local ISP for frame Relay service, and then connect the local local area network to the equipment provided by ISP (DCE) and "properly configure" it can be used. The purpose of this article is also to let you read it, can correctly achieve the "appropriate configuration" and understand the principle behind the configuration! As for the working mechanism and construction of the frame Relay network itself, it will not be discussed here because of its complexity.

Background:

For example, if you want to connect seven new remote sites from the company headquarters, but the router has only one free serial port, you can use frame Relay to save the scene! Of course, it should be pointed out that this can also lead to a single point of failure, which is not good. But frame Relay is used to save money, not to improve network resilience. (quoted from the CCNA Learning Guide).

Key features of frame Relay:

Frame Relay provides a connection-oriented transmission service

The data transmitted locally by the user will arrive at the destination through the network in order, and the peer does not need to reorder the received data when receiving the data. Before transmitting data, frame Relay will establish a logical path with each other through the network, which is called virtual circuit! So what is meant by "for connected services"? As mentioned above, the data has established a fixed connection before transmission, users always use this channel to transmit data, and the problem of information misordering will not occur. Corresponding to this is "connectionless network service", which means that without establishing a fixed connection before transmission, the data sent by the user may reach the end point through different ways (the result of routing). Therefore, the order of the data received at the destination may not be consistent with the order sent by the other party, so it needs to be reordered. The network built by routers belongs to this kind of network. As shown in the following figure:

There are two kinds of virtual circuits.

Virtual circuits are divided into permanent virtual circuits (PVC) and switched virtual circuits (SVC):

PVC is permanently established, similar to a dedicated line, defined by ISP with a static exchange table on a frame Relay switch, which exists all the time, regardless of whether the user transmits it or not.

SVC is temporarily established, it is dynamically established before transmission, and the line will be dismantled and disconnected after the transmission is completed. At present, the most common is to use PVC!

Frame Relay has two rates:

Access rate: this is the specific transmission rate of the line, which determines the speed of data transmission on the network. For example, the access rate of T1 is 1.544Mbps.

(CIR) commitment information rate: the highest average data transfer rate on a frame Relay circuit. It is usually slower than the transmission rate; when transmitting burst data, the transmission speed can exceed CIR. CIR controls your maximum transmission speed. As mentioned above, if the telecommunications network has sufficient resources, the maximum transmission speed that the customer can reach can reach the access rate (T1). On the contrary, if there are not enough resources, when the rate of CIR is exceeded, it may be discarded.

When users handle frame Relay services at ISP, they can choose a bandwidth according to their business needs, such as 512K, 1m, T1, 2m and so on. This bandwidth rate is CIR, which is the basic bandwidth rate guaranteed by ISP for users. But the transmission capacity of the virtual circuit adopts the dynamic on-demand allocation principle, during the transmission of data, the network will provide CIR, and according to the current network utilization, it is allowed to have a certain expansion. When the user has no data to send, the virtual circuit still maintains a connection, but the network can use the transmission capacity of the device as other services, so the virtual circuit will not monopolize the bandwidth (which is different from the dedicated line).

Frame Relay has congestion control capability

When the network is congested, the frame Relay switch sends congestion notifications to the devices at the sender and receiver, requiring the device to slow down the sending rate. If necessary, some packets that have been received will be dropped.

Frame encapsulation and frame format of frame Relay

Frame format 1 for frame Relay:

Frame format 2 for frame Relay:

Flag: indicates the beginning and end of the frame, with a value of 7e.

Frame Relay header: 16bit, address domain, for frame addressing.

Data: the length of the data carried by a frame can be varied, usually no more than 4096 bytes.

FCS: frame check sequence.

Header format of frame Relay:

Chand R: command / response (not used).

EA: 0 indicates that the address field does not end; 1 indicates the end of the address field.

FECN (forward display congestion advertisement): this message tells the router that the frame received has been congested on the path it is passing through.

BECN (backward display congestion Notification): this information is set on frames that encounter congestion, and these frames will be sent in the opposite direction to the congested frame. This information is used to help high-level protocols take appropriate actions when providing flow control.

DE: this information sets a level indication for frames indicating whether a frame can be discarded when congestion occurs. DE=1 indicates that this frame can be discarded first.

DLCI (data Link connection Identifier): total 10bit, used for addressing the target device. DLCI values range from 0 to 1007, of which 0 is used for signaling, 1 is reserved for 15, 16 is used for VC identification, and 992 is used for layer management.

There are two formats of frame Relay encapsulation: CISCO and IETF, which are slightly different and not compatible. The default encapsulation format of CISCO devices is CISCO, but it also supports IETF. Domestic frame Relay lines mostly use IETF.

Frame Relay addressing

The address used by frame Relay is a layer 2 address, called the data Link Identifier (DLCI). The total DLCI is 10bit, which is used for addressing the target device. DLCI values range from 0 to 1007, of which 0 is used for signaling, 1 is reserved for 15, 16 is used for VC identification, and 992 is used for layer management. The purpose of DLCI is to build virtual circuits, it is not unique, different users of frame Relay connections can use the same DLCI.

DLCI can be configured only if it is a device on the side of the ECE or if a subinterface is enabled. In the application of frame Relay, the DCE end is set on the ISP side, and the user's router belongs to the DTE segment, so if the user's router does not use subinterfaces, it is not necessary to configure DLCI.

You can view the local DLCI value with the show frame-relay map command. As shown in the figure above, on a DTE router, you can use a secondary command to check the local DLCI value. For example, 12.12.12.2 50 indicates that IP12.12.12.2 (opposite end) corresponds to DLCI 50 (local side). As long as the data sent to 12.12.12.2 is sent directly from the interface of DLCI 50, 12.12.12.1 80 means that as long as data is sent to 12.12.12.1, it is sent from the interface with DLCI 80. In fact, the correspondence between IP and DLCI is similar to that between IP and MAC.

Frame Relay Local Management Interface (LMI)

LMI is used between the local router and the first frame Relay switch to which it connects, allowing the provider network and DTE (local router) to exchange information about the operation and status of virtual circuits, including the following.

Survival message: verify that data is being transferred.

Multicast: this is an optional extension of the LMI specification that enables efficient distribution of routing information and ARP requests over frame Relay networks. Multicast uses the reserved DLCI-I019-1022.

Global addressing: make DLCI globally meaningful and frame Relay Cloud like LAN. So far, it has never been used in a production network.

Virtual circuit status: provides DLCI status. When LMI data streams are not sent periodically, status queries and messages are used as survival messages. After receiving LMI information from the service provider's frame Relay switch by using a frame Relay encapsulated interface, the router updates the status of the virtual circuit to one of the following three states:

Active status: everything is normal and routers can exchange information with each other.

Inactive: the router interface is in the up state and a connection to the exchange is established, but the remote router is in the down state.

Demolition status (deleted state): no LMI information was received from the switch, which may be caused by a mapping problem or a line failure.

LMI does not exchange information between customer routers, but between customer routers and the nearest frame Relay switch. Therefore, it is entirely possible that the router on one side of the PVC is receiving LMI information, while the router on the other end does not. Of course, the PVC does not work properly when one end is in the down state. The purpose of the above situation is to clarify the local characteristics of L-positive communication.

LMI provides the extended features of frame Relay, its main role is to provide frame Relay connection status detection services. LMI consists of multiple timers and counters, and there are currently three versions of LMI: ANSI, Q933a, and CISCO. The LMI type used by the user router must be the same as the LMI type of ISP. The default type of LMI for CISCO routers is CISCO. If the LMI of ISP is not CISCO, it needs to be changed to the effect type. In the early days, the router needed to specify the local LMI type manually, but now the router can automatically recognize the LMI type on the side of the DCE and automatically adjust it to the corresponding type. Manual configuration can be done with the following command:

Frame-relay lmi-type {ansi | q933a | cisco}

This is the end of the theoretical part. In fact, there are many details of frame Relay not mentioned here, interested friends can go to the Internet to search for in-depth research. The purpose of this article is to share frame Relay with you in an easy-to-understand, illustrated way so that beginners are no longer afraid of frame Relay! In the next part, we begin to introduce the configuration of frame Relay and routing protocols!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.