ATM can provide unprecedented scalability and cost performance, as well as support for future real-time services, multimedia services, etc. ATM will play an important role. But the current information system, namely LAN and WAN, is based on network layer protocols such as IP, IPX, AppleTalk, etc. Therefore, the key to the success of ATM and the development of the Internet is the interoperability of existing network technologies and ATM. The key to achieving this goal is that the same network layer protocols, such as IP and IPX, are applied to existing networks and ATMs at the same time, because it is the task of the network layer to provide a unified network perspective for high-level protocols and applications. So far, there have been many ways to run IP on ATM, such as: LANE and MPOA of ATM Forum, CLIP and NHRP of IETF, IP exchange of Ipsilon Network Company and Cisco's mark exchange, which will be introduced one by one.
1. Introduction
ATM and the existing protocol system, especially the IP and IPX protocols at the network layer, coexist for a long time. How to implement the existing network protocol and ATM on a single network, and how to interconnect ATM with the traditional network? The subject of research by a large number of researchers, designers and practitioners. However, ATM and IP originate from different technical groups and foundations and have their own applications. The purpose of IP is to send packets to the destination in an uncertain state. It is non-connected and there is no guarantee of quality of service. The purpose of ATM is to provide a guaranteed integrated service. It is connection-oriented and based on fast fixed-length letters. Yuan exchange. The huge difference between ATM and IP makes it difficult to effectively integrate the two.
There are two different models for supporting IP in an ATM network. These two models view the relationship between the ATM protocol layer and IP from different angles.
The first is the peer-to-peer model, which essentially considers the ATM layer as an IP peer layer. This model recommends using the same address scheme in the ATM network as the IP-based network, so the ATM endpoint will be identified by the IP address ATM signaling will carry such an address, and the routing of ATM signaling also makes the existing network layer routing protocol. Because the existing routing protocol is used, the peer-to-peer model eliminates the need to develop new ATM routes. The peer-to-peer model simplifies the address management of the end system, and greatly increases the complexity of the ATM switch, because the ATM switch must have the function of a multi-protocol router to support the existing address scheme and routing protocol. In addition, existing routing protocols are developed based on current LANs and WANs and cannot be well mapped into ATM and use the quality of service features of ATM.
In the current solution, IP switching and label switching are based on the peer-to-peer model.
The other model is called the subnet or overlay model, which separates the ATM layer from the existing protocol and defines a completely new address system, that is, the existing protocol will run on top of the ATM. This overlay model needs to define a new address system and related routing protocols. All ATM systems need to be given both an ATM address and the high-level protocol address it wants to support. The ATM address space is logically separated from the address space of higher-level protocols, without any correlation. Therefore, all protocols running on the ATM subnet require some kind of ATM address resolution protocol to map high-level protocol (such as IP) addresses to the corresponding ATM addresses. This method of separating ATM from high-level protocols allows for independent development, which is very important from a practical engineering perspective.
In the current solution, LANE, MPOA and CLIP are based on the coverage model.
Second, LANE
1. How to run IP on traditional LAN?
In a traditional LAN, when the source host wants to send a packet to a destination host on the same subnet, it checks its ARP cache to see if it already knows the hardware address (MAC address) associated with the destination host ’s IP address. If it already knows, then Send the packet with the IP address and MAC address of the destination host.
If the destination MAC address is unknown, the source host sends an ARP request packet. The ARP request is a local broadcast packet that will be received by all hosts in the subnet. After the destination host recognizes its own IP address, it responds to its MAC address in the ARP response packet. The source host receives the ARP response and stores it in its own ARP table. Now the source host can send the packet with the correct destination IP address and MAC address.
2. What functions must the ATM LAN emulate?
(1) Because the traditional LAN is a media sharing network, it is easy to provide broadcast services and implement ARP. The ATM network must imitate this function and be implemented by BUS (broadcast and unknown server).
(2) Generally speaking, each host in a traditional LAN has its MAC address and IP address. In addition to the ATM address, the host directly connected to the ATM network must also have a MAC address and an IP address.
(3) The ATM host must provide the same service as the interface service provided by the MAC protocol to the network layer protocol, such as the NDIS or ODI type driver interface.
3. How does LANE work?
As the name implies, the function of LANE is to emulate LAN on an ATM network. The LANE protocol defines a mechanism to emulate IEEE 802.3 Ethernet or 802.5 token ring network. The LANE protocol defines the same interface as the service provided by the existing LAN to the network layer, and the data transmitted in the ATM network is encapsulated in the corresponding LAN MAC packet format.
Each ELAN (Emulated LAN) consists of a group of LANE customers (LEC) and LANE services. LEC can also be a bridge and a router acting as an ATM host agent. The LE service consists of three different functional entities: LAN emulation configuration server (LECS), LAN server (LES) and BUS. These three service entities can exist separately, but they are usually located on the same device. For example: LES can be located on ATM switches, Routers, bridges and workstations.
The following are the steps for communication between a workstation in LANE and another workstation:
(1) Initialization
The LEC needs to know the LECS ATM address and establish a connection with it. This is done through the ILMI or well-known LECS address. At any point in the process, the LEC can establish a two-way configuration with the manually configured LECS address directly to the VCC. During this process, LEC will obtain the ATM address of the LES of the ELAN.
(2) Registration This is the mechanism by which LEC provides address information, such as MAC address, to LES. In this process, a pair of connections will be established between LEC and LES, that is, two-way point-to-point control directly to VCC, and one-way point-to-multipoint control distributed VCC.
(3) Address resolution This is how LEC learns the ATM address of the destination site from LES. It is implemented by the ATM address resolution protocol, which allows LEC to establish data directly to VCC to transmit frames. At this time, a two-way point-to-point multicast transmission VCC and a one-way point-to-multipoint multicast forwarding VCC are established between LEC and BUS.
(4) Data transmission During the transition period when the source site and the destination site are waiting to establish a data direct VCC, BUS can forward the frame to all LECs in the ELAN. When the data direct VCC is established, the communication is from the original route (BUS) To switch to a new route, in order to ensure the order of the frames, the flush message protocol is used to notify the BUS: when the new route is used to transmit the frame, a clear request is sent to the BUS and forwarded to all LECs in the ELAN , And then no longer frames through BUS (old route), all frames will be sent to the destination site through data directly to VCC (new route).
It should be noted that in the above process described in the ATM Forum specification, there is no mention of the resolution from the IP address to the MAC address. The following is the whole process of communication between the host of the traditional LAN and the ATM host:
(1) To determine the MAC address of the destination site, the source host broadcasts an ARP request containing an IP address. This is a standard process for any IP network. The ARP request reaches the LAN / ATM bridge on a traditional LAN.
(2) The LEC on the LAN / ATM bridge forwards the broadcast packet to the BUS via multicast transmission VCC, and the BUS forwards the VCC to all members in the ELAN via multicast to send an ARP request.
(3) The destination site receives the ARP request and recognizes its own IP address. In response, it puts its MAC address in the ARP response. Because this is not a direct VCC to the LAN / ATM bridge, the LEC at the destination site sends the ARP response to the BUS via multicast sending VCC, and the BUS forwards it to the LAN / ATM bridge via multicast sending VCC.
(4) The LAN / ATM bridge transmits the ARP response to the source host through the traditional LAN.
(5) At this time, the source host has the MAC address of the destination site and starts to transmit data through the LAN.
(6) The bridge sends the VCC through multicast to transmit the packet to the BUS, and the BUS forwards the packet to the destination site.
(7) At the same time, the LEC on the LAN / ATM bridge sends a LE-ARP request to the LES through the control direct VCC, asking for the ATM address corresponding to the MAC address of the destination site. If the LES does not have the mapping, the VCC is distributed through the control. All LECs send LE-ARP requests. After receiving the request, the destination site LEC puts its ATM address in the LE-ARP response and sends it back to the LES through the control directly to VCC.
(8) The source LEC receives the LE-ARP response from the LES by controlling the direct VCC, extracts the ATM address and establishes a data direct VCC between the source and destination.
(9) After the data direct VCC is established, the packets from the bridge will be transmitted through the data direct VCC instead of BUS.
4. The advantages and limitations of LANE
Because LANE provides the same service interface as the driver provided by the existing MAC protocol to the network layer, there is no need to change the driver, which will accelerate the development and application of ATM. However, the function of LANE is to make the characteristics of ATM transparent to high-level protocols, so it prevents high-level protocols from taking advantage of the inherent advantages of ATM, especially its quality of service guarantee. The newly completed version 2.0 of LANE provides locally managed quality of service for communication between ATM end systems. The protocol provides a mechanism to determine whether to support the desired quality of service. Each locally defined quality of service may contain information to indicate whether the VCC established with the quality of service can be shared by other protocols or applications.
Although LANE provides an effective way of bridging within ATM network subnets, services between subnets still need to be forwarded through routers. Therefore, ATM routers are likely to become a bottleneck. The MPOA discussed below will solve the problem of efficiency of communication between subnets.
3. CLIP (Classical IP over ATM)
1. Principle
To run IP on an ATM network, IETF uses the concept of logically independent IP subnet (LIS). Like a normal IP subnet, a LIS contains a set of IP nodes (such as hosts or routers) connected to a single ATM network, and they belong to the same IP subnet. ATM LIS behaves much like a traditional IP subnet. In order to resolve the address of a node within the LIS, each LIS provides an ATMARP server. All nodes (LIS clients) in the LIS are configured with the ATMARP server's ATM address. When a node in LIS appears, it first establishes a connection with the ATMARP server. Once the ATMARP server detects a new LIS client's connection, it sends a reverse ARP request to the client, asking for the node's IP address and ATM address, and saves it in its ATMARP table. Subsequently, any node in LIS that wants to resolve the destination IP address will send an ATMARP request to the server. If the address mapping is found, the server returns an ATMARP response. Otherwise, it returns an ATM_NAK response to indicate that there is no such mapping. The mapping table unless the client responds to its periodic reverse ARP request. Once the LIS client obtains the ATM address corresponding to the IP address, it can establish a connection with the address. Protocols for packet encapsulation and address resolution are defined in RFC1483 and RFC1577, respectively.
However, because the address resolution protocol defined in RFC1577 retains the host's requirement to send packets to sites outside the subnet through the default router, the shortcut VCC can only be established between nodes in the same subnet, otherwise the source site must group the packets Forward to the default router, even if the source and destination sites are in the same ATM network. In this way, the ATM router becomes a bottleneck, and the quality of service cannot be achieved.
Compared with LANE, RFC1577 only supports IP and does not support other network layer protocols such as IPX and AppleTalk. In addition, CLIP does not support multicast, which is also an important shortcoming of RFC1577.
2. Expansion of CLIP
2.1, NHRP (Next Hop ResoluTIon Protocol)
In order to provide shortcut routing between sites in the same ATM network and different subnets, IETF proposed a protocol called NHRP, which is built on the CLIP model, but replaced LIS with the concept of non-broadcast multiple access network (NBMA) The concept of NBMA means that multiple devices can be connected to the same network, but can be configured to different broadcast domains, and supports direct communication between hosts in different LIS. Frame Relay and X.25 are examples of NBMA networks.
NHRP replaces the ARP server with the concept of NHS (NHRP server). Each NHS contains a "next hop resolution" cache table, which contains the IP to ATM address mapping of all nodes related to the NHS. The node configures the ATM address containing the NHS, and registers its ATM address and IP address in the NHS with a registration packet.
The protocol process is as follows: When a node wants to send a packet through the NBMA network, that is, when a specific ATM address needs to be resolved, it generates and sends an NHRP request packet to the NHS. Such a request and all NHRP information are sent through an IP packet. If the destination site is served by the NHS, the NHS returns its address through the NHS response packet, otherwise the NHS looks up its routing table to determine the next NHS to reach the destination and forward the request. The same algorithm is executed at the next NHS until the NHS of the requested mapping is truly known. The destination node returns an NHRP response, goes through the same series of NHS in the reverse order, and reaches the requesting node. The requesting node can establish a direct data connection. Thus, the ATM VCC can be established across the subnet boundary, so that the subnets can communicate without routing.
2.2, Multicast
There are two ways to support multicast.
The first is through a multicast server, all nodes that want to send multicast information establish a point-to-point connection with it, and it connects with all receiving nodes through a point-to-multipoint connection. The multicast server receives data through the point-to-point connection and retransmits the data through the point-to-multipoint connection. This method can be used for large networks, but the multicast server may eventually become a bottleneck.
The second method is called a multicast network. Each node in the group establishes a point-to-multipoint connection with other nodes. In this way, all nodes can send and receive data from other nodes. For a group with N nodes, N point-to-multipoint connections will be required, which is not suitable for groups with a large number of nodes.
Both methods are used in the Multicast Address Resolution Server (MARS) proposed by Armitage. MARS serves a cluster of nodes, and all end systems in a cluster are configured with MARS ATM addresses. When an end system wants to send a message to a specific multicast group, it establishes a connection with MARS and sends out a MARS_REQUEST message. MARS returns MARS_MULTI message. This message contains the address of the group ’s multicast server or the address of a group member. Supports a multicast server, the requesting node establishes a connection with the server, sends data to the server, and the server forwards the data to the nodes in the group; in the multicast network scheme, the requesting node establishes a point with the nodes in the group Go to a multipoint connection and send data over that connection.
4. MPOA
1. The principle of MPOA
The purpose of MPOA is to effectively transmit unicast data between subnets in a LANE environment. MPOA integrates LANE and NHRP to retain LANE, while improving the efficiency of communication between subnets by bypassing routers. MPOA allows physical separation of network layer routing calculation and data transmission, which is called virtual routing. The routing calculation is performed by the server located in the router—namely MPS—and the data transmission is performed by the client located in the edge device—namely MPC.
At the entry point, MPC detects the data flow transmitted to the router containing MPS through ELAN. When it finds a shortcut that can bypass the current routing path, it uses the NHRP-based protocol to request a shortcut to the destination node. This information is recorded in its entry table, a shortcut VCC is established, and frames are sent through the shortcut VCC. For packets using shortcuts, MPC removes the data link layer (DLL) encapsulation from the packet.
At the exit point, the MPC receives network data from other MPCs. For frames received via shortcuts, the MPC plus appropriate DLL encapsulation transfers them to the upper layer protocol. The DLL package information is provided by MPS and stored in the export buffer.
MPS is the logical component of the router and provides MPC with network layer forwarding information. It contains the complete NHS defined in NHRP. The MPS interacts with the local NHS and routing functions to answer the MPOA request of the ingress MPC and provide DLL packaging information to the egress MPC.
The following is a brief description of the communication process within and between ELANs.
Intra-LAN communication from one MPOA host or LAN host to another MPOA host or LAN host in the same ELAN. These data streams use ELAN for address resolution and data transmission. For communication between ELANs, from one MPOA host or LAN host to a MPOA host or LAN host of different ELAN, the short path uses the default path, the long path uses the shortcut, the default path uses ELAN and the router, and the shortcut uses LANE and NHRP Address resolution and shortcuts. The shortcut works like this: If the source node and the destination node are not in the same MPS management domain, the ingress MPS translates the MPOA resolution request into an NHRP resolution request, and forwards the request to the egress MPS through the NHRP. When the egress MPS receives the egress MPC ’s After the response, it generates the NHRP parsing response and sends it back to the ingress MPS. When the ingress MPC gets the MPOA parsing response from the ingress MPS, a shortcut can be established between it and the egress MPC.
2. The advantages and limitations of MPOA
MPOA fundamentally separates data transmission and route calculation, and distributes functions to different devices, thereby reducing the number of devices involved in route calculation and the complexity of end devices. It can support Layer 2 and Layer 3 network interconnection in a unified manner, thus ensuring large-scale interconnection in the ATM environment. It can effectively handle burst data and long-term data streams at the same time, but the complexity of MPOA is very controversial.
Five, IP exchange
The purpose of IP switching is to obtain the most effective IP implementation on fast switching hardware, complementing the advantages of non-connected IP and connection-oriented ATM. IP switching is a standard ATM switching plus an intelligent software controller connected to the ATM switch port, that is, an IP switching controller. The IP switch hands the initial packet of the data flow to the standard routing module (part of the IP switch) for processing. When the IP switch sees enough packets in a flow and considers it to be long-term, it is the same as the adjacent IP switch or edge device With the establishment of flow labels, subsequent packets can be label switched at high speed, bypassing the slow routing module. Special IP switching gateways or edge devices are responsible for the conversion from unmarked packets to marked packets and packets to ATM data.
Each IP switching gateway or edge device that connects an existing network device to an IP switch establishes a virtual channel to the IP switch controller as a default forwarding channel at startup. When receiving a packet from an existing network device, the edge device The packet is transferred to the IP switching controller through the default forwarding channel.
The IP switch controller implements traditional routing protocols, such as RIP, OSPF, and BGP, and forwards the packet to the next node through the default forwarding channel in a normal manner. This may be another IP switch or edge device. The IP switch controller also performs data flow classification, which recognizes long-term data flows, because such data can be optimized by ATM hardware's cut-through exchange, and the rest of the communication still uses the default method, point-to-point storage Forward routing.
When a long-term data flow is identified, the IP switching controller requires that it be marked in the previous section and use a new virtual channel. If the source edge device agrees, the data flow will flow to the IP switching controller through the new virtual channel. The next node also performs the same action. When the flow uses special input channels and output channels independently, the IP switch controller instructs the switch to establish appropriate hardware port mapping, bypassing routing software and related processing expenses. This process continues. The first few packets of the flow establish a direct connection from the source edge device to the destination edge device. This design enables IP switches to forward packets at a rate limited only by the switching engine. The first generation IP switches support throughputs of up to 5.3M packets per second. In addition, because there is no need to encapsulate ATM cells into IP packets of the intermediary IP switch, the throughput in the IP network is also optimized.
Ipsilon proposed two protocols to the IETF. The General Switch Management Protocol (GSMP, RFC1987) allows the IP switch controller to access the switch hardware and dynamically change the switching mode: storage forwarding or cut-through. The Ipsilon traffic management protocol (IFMP, RFC1953) is used to exchange control information between the edge device and the IP switching controller and associate the IP flow with the ATM virtual channel.
An important feature of IP switching is that the classification and switching of flows is performed locally, rather than on an end-to-end basis. This preserves the disconnected nature of IP and allows IP switches to bypass failed node routes without the need to restart from the source host. Establish a channel.
In addition, flow classification makes IP switching equally effective in supporting long-term and burst data.
However, IP switching is based on flows, and its scalability is questionable in large networks. In a large network, the number of flows may eventually exceed the number of available virtual channels.
Five companies officially declared support for Ipsilon's IP exchange: Ericsson, General Datacomm, Hitachi America Ltd., NEC America Inc. and DEC Ipsilon. They tried to make this technology the de facto standard-MPLS.
Six, mark exchange
Another option is Cisco's tag exchange. The label switching network contains three components: a label edge router, a label switch, and a label distribution protocol.
Mark edge routers are routing devices with complete layer 3 functions at the edge of the mark-switched network. They inspect incoming packets and mark them appropriately before forwarding them to the mark-switched network. When the packet exits the mark-switched network, the mark is deleted. As a router with complete functions, the labeled edge router can also apply value-added layer 3 services such as security, billing, and QoS classification. The ability to mark edge routers does not require special hardware. It is implemented as an additional feature of Cisco software. The original router can be upgraded with software to have the function of marking edge routers.
The label switch is the core of the label switching network. The so-called tag is a short, fixed-length tag, so that the tag switch can use fast hardware technology to do simple and fast table query and packet forwarding. The label can be located in the VCI field of the ATM cell, the flow label field of IPv6, or between the Layer 2 and Layer 3 header information, which makes label switching available on a wide range of media, including ATM connections, Ethernet, etc.
The label distribution protocol provides a method for the label switch to exchange label information with other label switches or label edge routers. Marked edge routers and marked switches use standard routing protocols (such as BGP, OSPF) to establish their routing databases. Adjacent tag switches and edge routers distribute tag values ​​stored in the tag information base (TIB) to each other through a tag distribution protocol.
The following is the basic processing of the label switching network.
(1) Marked edge routers and marked switches use standard routing protocols to identify routes, and they are fully interoperable with non-marked routers.
(2) The label edge router and switch assign label information to the routing table generated by the standard routing protocol and distribute it through the label distribution protocol. The label edge router receives the label distribution protocol information and establishes a forwarding database.
(3) When the label edge router receives the packet that needs to be forwarded through the label-switched network, it analyzes its network layer header information, performs available network layer services, selects a route for the packet from its routing table, marks it, and forwards it to the next A node's label switch.
(4) The tag switch receives the tagged packet and only exchanges based on the tag, without analyzing the network layer header information.
(5) The packet reaches the mark edge router of the exit point, the mark is stripped, and then it is forwarded.
In a label-switched network, label distribution protocols and standard routing protocols can be aggregated with a target prefix labeling algorithm, which can establish labeling information in the TIB before the data flow passes through the network. This has two meanings. One is that all packets in the stream can be label-switched, even for short bursts of data; in addition, it is based on topology, with a label assigned to each source / destination. In IP switching, only after a certain number of packets have passed by a long-term data stream, a shortcut is established. Therefore, label switching uses labels more effectively than flow-based mechanisms, avoiding the process of establishing a stream one by one, which gives it the good scalability required by public Internet service networks. In the public Internet, the number of streams It is huge, and its rate of change is also very high.
Other vendors have similar mechanisms, such as Cabletron's SFVN (Secure Fast Virtual Networking), Cascade's IP Navigator, DEC's IP packet switching, Frame Relay Technologies' Framenet Virtual WAN switching, and IBM's ARIS (Aggregate Route-based IP Switching). .
7. Conclusion
This article briefly introduces some programs that support IP on the ATM network. These programs are based on the assumption that the traditional LAN and router are connected through the ATM network, or that the hardware platform is the ATM network, and the application is based on IP. Other contents will not be introduced here.
Waterproof RGB LED Dance acrylic Floor Display Screen for Wedding Party
These interactive display walls are applied in different purposes. This is a perfect audio/visual solution for commercial or
business purposes as well as for the educational purposes. It is often used for providing information to the viewers or audiences in the most interactive and effective way .
prospective customers. The interactive LED can be used IN retail stores, restaurants and other commercial spaces for giving information about their services and products.
1.Aluminum Structure Light Weight,good heat dissipation, good weather resistance, convenient transportation.
2.Imported PC Protrctive Surface, High Permeability, Strong Impact Resistance, Good anti-slip Effect
3.Strong Load Capacity The center of the panel has a load-bearing pivot,with a single panel bearing more then 1.2tons
Dance LED Display,Flexible Led Screen,Indoor Led Display Screen,Led Church Screen
Guangzhou Chengwen Photoelectric Technology co.,ltd , https://www.cwledpanel.com