Application of ATM technology in computer Communication Networks

The computer communication network involves two fields of communication network computers. The communication network provides necessary means for data transmission and exchange between computers. Computer technology permeate the communication network and improves the performance of the latter. As a mature technology, IP plays a positive role in the rapid development of the Internet based on the features of network availability, best effort, and client-centric. As a modern communication technology, the application of ATM in computer communication networks is also in line with the future development trend of telecommunication networks and information networks. It is characterized by reliable, manageable, and road-centered networks: digitalization, broadband, and segmentation. Both technical factors and market demands are driving the development of ATM. In the face of the rapid development of the IP service, the ATM, known as "aristocratic technology" in the industry, can be integrated into the computer communication network only by actively seeking the best solution for ATM to support IP.
I. cell structure and principle of ATM
ITU defines ATMAsynchronous Transfer Mode): ATM is an asynchronous Transfer Mode. Asynchronous refers to the nature of ATM statistics reuse. The transfer mode refers to the multiplexing, exchange, and transmission technologies used in the network, that is, the transfer of information from one place to another. In this transfer mode, information is organized into a CELL), and each CELL from a user information does not need to appear periodically. Therefore, ATM is a technology that performs statistics multiplexing, switching, and transmission in units of cells in the network.
ATM is a new type of grouping technology. A cell is actually a group with a fixed length. The total length of a cell is 53 bytes, 5 of which are header and 48 are information segments, or net load. The header contains control information such as the logic address and priority level that indicate the target of a cell. Information segments load information from different users and services. Information of any business is cut and encapsulated into cells in a unified format.
ATM uses the asynchronous time division and repeat method to bring together cells from different information sources and queue in the buffer. The cells in the queue are output to the transmission line one by one based on their arrival priorities, form a cell stream that is connected at the beginning and end. Cells with the same sign do not correspond to a fixed time slot on the transmission line, nor do they appear periodically. Asynchronous time division multiplexing makes ATM flexible. Any service occupies resources based on the actual amount of information, so that network resources can be used to the maximum extent. In addition, no matter how different the nature of the business source is, such as the speed, burstable size, quality, and real-time requirements), the Network is handled in the same mode, truly achieving complete business integration. To improve processing speed, ensure quality, reduce latency and loss rate, ATM works in connection-oriented mode. When the communication starts, a virtual circuit is established, and the virtual circuit mark is written into the header. The network sends the cell to the destination according to the virtual circuit mark. Virtual circuits can be removed and released. On an ATM node, only virtual circuit switching is completed. To simplify network control, the ATM assigns Error Control and traffic control to the terminal, and does not need Error Control and Traffic Control for segment-by-segment links. Therefore, ATM combines the advantages of circuit switching and group switching, that is, it takes into account both the Statistical Multiplexing of the group switching mode, the flexibility and efficiency, the small transmission delay of the circuit switching mode, and the good real-time performance. It can carry a variety of information media in a single main network, carry a variety of communication services, and ensure QoS. There are two types of ATM switching: VP switching and VC switching. VP switching means that only the VPI value is changed during the switching process, and the VCI value is passthrough, while VPI and VCI change during the VC switching process.
Ii. ATM Protocol Reference Model and layer Functions
The ITU-T Protocol Reference Model is defined in the I .321 recommendation of B-ISDN, which is a three-dimensional model, including three planes: User plane U, control plane C and management plane M, each plane is layered, including the physical layer, ATM layer, AAL layer, and high layer.
The three aspects of the Protocol Reference Model provide different functions:
1. User plane U: provides the user information stream transmission function, and also has certain control functions, such as traffic control and error control;
2. Control Plane C: Provides call control and connection control functions, and establishes, monitors, and releases calls and connections using signaling;
3. Management plane M: provides two management functions: Layer Management and face management. Layer Management Layer) to complete management functions related to the resources and parameters of various protocol layer entities, such as meta-signaling, and also process OAM information flows related to each layer; surface management is not hierarchical ), it completes management functions related to the entire system and coordinates all planes.
For ATM protocols, refer to the functions of each layer in the model:
1. The physical layer is divided into two sub-layers: PM physical media sub-layer) and TC transmission convergence sub-layer ). PM physical media sub-layer) is responsible for line encoding photoelectric conversion and bit timing to ensure correct transmission of data bit streams. The feature of the transfer convergence sub-layer is cell rate decoupling and HEC generation/verification; cell demarcation; transmission frame adaptation; transmission frame generation/recovery.
2. the ATM layer provides four functions: general traffic control, header generation and extraction, cell VPI/VCI translation, and cell multiplexing and routing.
3. The AALATM adaptive layer is used to adapt high-level functions to ATM cells. The purpose of the AAL layer is to enable different types of services, including management plane and control plane information, to be transmitted in a unified form of ATM cells after adaptation. The AAL layer is directly related to the business. The AAL layer adapts to different types of services. For ATM users, AAL is implemented in user terminal devices; for non-ATM users, AAL is implemented in UNI network devices. The AAL layer is divided into two sub-layers: The disassembly sub-layer SAR and the aggregation sub-layer CS. At the sender end, the business flow must be adapted to the ATM layer. SAR segments the high-level information into fixed-length and standard-format ATM cells. At the receiver end, when forwarding the information at the ATM layer to the upper-layer, SAR receives ATM cells and reassembles them into high-level protocol information formats. CS implements the transmission of scheduled information, error detection and processing, processing of cell transmission delay, and identification and processing of user data units.
Iii. simple comparison of ATM and IP technology
1. Service Quality Assurance
QoS is a sign of service reputation for operators and users. Due to the distinction between connection-oriented and non-connection-oriented, IP technology and ATM technology have fundamental differences in QoS Assurance. The length of the IP package is not fixed. The delay between the long information package and the short message package varies greatly. This introduces a large delay jitter and is not suitable for real-time services. When users increase, the service quality decreases, resulting in unstable service quality. Currently, some real-time applications use low bandwidth utilization in exchange for high service quality. The fixed-length cells are used in the ATM technology to achieve the same packaging and package splitting latency, reduce latency jitter, and reduce the delay value with a small cell length. In addition, ATM uses traffic control technology. Before the connection is established, it uses signaling negotiation to determine whether the customer's service quality requirements can be guaranteed. Only after the network is confirmed can the customer accept the incoming traffic, ensure different service quality for each virtual circuit. This is the true quality of service (QoS ). To make up for the defects of IP technology in terms of service quality, TETF proposes a Protocol such as RSVPResource Reservation Protocol. However, due to the limitations of the IP address itself, the protocol implementation is complex and the progress is slow.
2. Protocol simplification
From the perspective of the development process, lP technology was not ideal in the early stages of development, so it adopted technology such as segment-by-segment error correction and feedback retransmission, which made the Protocol complex. With the development of optical fiber technology and the improvement of transmission and processing capabilities, these parts have become redundant. However, due to investment in software and hardware, IP itself becomes an obstacle and can only retain these features. This will inevitably affect the network transmission efficiency. The ATM technology makes full use of the development of the optical fiber technology, simplifies Error Control, does not check service integrity at the intermediate node, and greatly simplifies the protocol. ATM Traffic Control and signaling are used for precise control to improve network utilization and ensure service quality. Although complicated, it is worth it.
TCP/IP is the basic protocol of the Internet. The IP layer in the TCP/IP framework corresponds to the network layer in the OSI reference model to complete routing selection and group forwarding, TCP corresponds to the transport layer in the OSI reference model to complete end-to-end data collection confirmation and error correction. It can be seen that the IP protocol is essentially a data protocol that does not need to establish a connection in advance, but directly depends on the IP grouping header information to determine the packet forwarding path. Technically speaking, it has the following features: 1. distributed structure; 2. End-to-End principle. All value-added functions are completed by terminals outside the network; third, the IP network can be built on any transmission channel to ensure interoperability between different network types. Fourth, it has a unified addressing system and high network scalability.
ATM works in connection-oriented mode. When communication starts, a virtual circuit is established, and the virtual circuit mark is written into the header. The network sends the cells to the destination according to the virtual circuit mark. To simplify network control, the ATM assigns Error Control and traffic control to the terminal, and does not need Error Control and Traffic Control for segment-by-segment links. It can improve processing speed, ensure quality, and reduce latency and loss rate. ATM combines the advantages of circuit switching and group switching, that is, it has the advantages of statistical multiplexing, flexibility and efficiency, low transmission latency, and good real-time performance. It can carry a variety of information media in a single main network, carry a variety of communication services, and ensure Qos.
Compared with ATM technology, IP technology is inherently inadequate in terms of service quality assurance, security, statistics reuse, traffic management, and congestion control. When the number of users is small, these problems are not so prominent. Once the number of users increases rapidly, the online data traffic increases, and the problem arises. Therefore, IP technology is inferior to ATM technology in terms of service quality assurance.
Iv. Application of ATM technology in computer communication networks: LANE
1. Concept of ATM LANE
As the name suggests, LANEEmulated LAN is used to simulate a LAN on an ATM network. The LANE Protocol defines a mechanism to simulate an IEEE 802.3 Ethernet or 802.5 ring network. The LANE Protocol defines the same interface as the service provided by the existing LAN to the network layer. The data transmitted in the ATM network is encapsulated in the corresponding lan mac group format.
Simulate a traditional LAN on the ATM network, and connect multiple traditional LAN and terminal devices through the ATM network. Create a new LAN on the ATM network, and the communication behavior between these LAN points is exactly the same as that of the traditional LAN. LAN simulation hides the ATM switching structure in the LAN, and the LAN terminal does not feel the existence of the ATM network. Therefore, the advantages of the ATM network can be used without modifying the hardware and software of the terminal device. It can be seen that the simulation protocol is mainly implemented on the ATM host and the atm lan bridge. An atm lan Bridge is a converter between a LAN and an ATM. It uses the ALL5 protocol to adapt LAN data, generate an ATM cell, or reorganize an ATM Cell to restore the data frame of the LAN. The LAN simulation function is added between the ATM Adaptation Layer and the high-level protocol to enable the ATM host to simulate the behavior of traditional LAN devices and communicate with the LAN.
The atm lane is designed for LAN access, that is, the LAN Emulation of the ATM ). For traditional IP terminals, the ATM network is like a LAN, which contains several IP subnets connected by routers.
LANE is based on the Client/Server mode). a lane Server can be connected to multiple LANE clients. LANE supports multiple protocols (MPOA: Multi-Protocol over ATM) for transmission, and allows interconnection between different lanes. LANE fully supports connectionless features in the LAN; LANE supports unicast, multicast, and broadcast transmission;
Advantages of atm lan:
Real-time information transmission. Because of the low transmission and exchange latency of ATM, it can ensure real-time information transmission. It has strong network processing capabilities, various services, including voice, data, and images, can be uniformly converted into ATM cells for transmission and processing in the ATM network. The transmission rate is high and the communication between the LAN and the public network is easy;
LAN simulation provides two layers of OSI-RM layer function, that is, using an ATM network to simulate LAN data link layer and physical layer function, and has nothing to do with all the high-level business, protocol and application. ATM switches do not directly participate in LAN simulation, but only provide virtual channels for ATM cells.
2. LAN simulation structure
The LAN simulation is logically composed of a server and a group of LAN simulation customers.
The LAN simulation server can be divided into three functions:
A. broadcast and Unknown address Server BUS: Broadcast & Unknown Server) implements Broadcast and multi-point communication functions. It not only simulates the Broadcast mechanism of the traditional LAN, but also establishes the LEC data unicast before the direct link between LEC, one LANE has only one BUS;
B. LAN simulation Server LES: LANE Server) is an address Server that stores the MAC address of each LEC in LANE.
The ing table with the ATM address to convert the MAC address and the ATM address. It provides the registration and Resolution Methods for the MAC address and responds to the preceding request of LEC; there is only one LES in a LANE ;)
C. LAN simulation structure Server LECS: LANE Configuration Server) There can be multiple lanes in the ATM network. LECS stores the structure information of the LANE and configures LEC to the LANE; maintains the configuration information of LEC, LES, and BUS in Multiple lanes in an ATM network; provides each LEC with its own les atm address)
LEC of LAN simulation can be implemented on an ATM workstation and an ATM bridge. Its main function is to convert MAC frames and ATM cells in a LAN; save the MAC, ATM, and VPI/VCI ing tables, complete the address resolution function together with the LAN simulation server, and start the signaling to establish an ATM virtual connection to provide a channel for data transmission; it can be used with the management interface of the ATM layer to implement management functions. When simulating Ethernet or ring-based network nodes on the ATM terminal system, at least one MAC address must be bound. The function is to encapsulate IP data packets and send them to the ATM network for transmission, translate the ATM group to re-form the IP datagram .)
Each LANE consists of a set of LANE customer LEC) and LANE Services. LEC can also be a bridge and router used as an ATM host proxy. The LANE service consists of three different functional entities: the LAN simulation configuration server LECS), the LAN server LES), and the BUS. These three service entities can exist separately, but are usually located on the same device, for example, LES can be deployed on an ATM switch, router, bridge, and workstation.
LECS, LES, and BUS are concentrated on the ATM workstation. LEC1 connects Ethernet to the ATM network, LEC2 connects the licensing ring network to the ATM network, LEC3 connects the FDDI network to the ATM network, and LEC4 is the ATM host. LUNI is the network interface of the LAN simulation user. LEC and server are connected through LUNI and ATM. The four LEC and the server form a simulated lan lane.
V. Application of ATM technology in computer communication networks II: DSLAM
DSLAMDigital Subscriber Liner Access Multiplexer) is defined as a digital user line Access multiplexing, local device, to complete the convergence and distribution of ADSL Access services.
The data signal uploaded by the atm dslam principle is in the ADSL Modem, and the conversion between MAC frames and ATM cells is realized through the AAL5 segmentation and sub-layer SAR) function. After DSL modulation, the signal is transmitted to the DSLAM side of the local terminal. The local DSLAM is demodulated and restored to the ATM Cell format. The atm dslam processes many of these upload signals and reconnects or exchanges the ATM signals according to certain conditions. At the same time, various ATM businesses are classified and managed to achieve a constant rate of CBR), real-time changeable rate xt-VBR), quasi-real-time variable rate nrt-VBR), and indefinite ratio UBR) and set different traffic control parameters and priorities for different types of ATM services to ensure the support of users with different requirements for high-performance QoS, provides permanent virtual circuit PVC) and establishes a connection with the upper-layer ATM switch. When the signal is run, the sent signal is transmitted to the atm dslam in the form of an ATM cell. These received ATM signals carry the relevant ATM information, for example, the ATM service type and PVC data have other parameters. Different service parameters are identified by ATM and different processing measures are taken, including the packet discard principle, cache mechanism, and traffic control mechanism, to ensure QoS of different services, then it is divided into the corresponding ADSL signal modulation, and then output to the peer ADSL Modem.
The current DSLAM is divided into ATM-DSLAM and IP-DSLAM. The main difference between the two is: ATM-DSLAM is the full ATM connection, while the IP-DSLAM is to connect the ATM to the DSLAM device related function board Ethernet Board), to achieve the end of the ATM, then, the ATM cells are converted to Ethernet frames for transmission. From the network mode: ATM-DSLAM and IP-DSLAM are used as access equipment, but the ATM-DSLAM is connected to the ATM convergence switch, and the IP-DSLAM is connected to the layer-3 switch. From the business application perspective: ATM-DSLAM application of PVC tunnel mechanism, security mechanism and bandwidth commitment mechanism, QoS performance is also good, suitable for open VPN business. Due to the termination of the terminal device, the ip dslam pvc has inherent limitations in data security and bandwidth control, and poor QoS performance. Therefore, it is more suitable for access-type Internet services. From the perspective of user authentication: ATM-DSLAM authentication is achieved by ending the PVC in the ATM Broadband Access Server BAS. The IP-DSLAM PVC ing will VLAN identifier ID) to the IP-BAS to authenticate the user that is to say if there is no vlan id to the upper-layer BAS, can not authenticate the user ). Therefore, the ip dslam must support a large number of VLANs, so that one-to-one ing and unique use of VLAN IDs on the PVC and VLAN IDs can represent users.
Based on advanced ATM technology, the atm dslam device fully supports a variety of ATM services, such as CBR, rt-VBR, nrt-UBR, and UBR, different processing measures, different cache space and scheduling priorities, and different service shaping functions are provided for these services, so that different ATM service types can be transmitted throughout the service, this ensures the bandwidth and high-performance QoS support. ATM is a connection-oriented transmission mechanism with a natural security isolation mechanism, which is PVC and therefore prevents information leakage.
2. DSLAM Application
DSLAM applications have a concise network topology structure, which may vary with the user's needs, networking conditions, and the distribution range of local DSLAM-to-user (ADSL/VDSL) Modem twisted pair cables, select ADSL/VDSL access. In addition to access modes such as ADSL and VDSL, the ATM-based DSLAM also supports leased line services such as CES and FR, which is suitable for multi-service access to applications.
DSLAM is used as the access layer device to converge to the terminal ATM switch through the ATM port, and enters the ATM Backbone Network through the man ATM switch, providing users with Internet access and various leased line services.
For different types of ADSL Access Users, DSLAM works with other devices, such as BASBroadband Access Server, to provide flexible and variable billing methods:
1) for common home users, PPPoE is used for access to the BAS, and BAS, the Radius server and the Billing server are used together to complete user authentication, traffic management, and Billing.
2) for SOHO users, configure PPPoA and NAT on the Modem. The whole office shares an account, and computers in the SOHO access the Internet and other services through the Modem. The Modem initiates a PPPoA call to BAS. The service selects the gateway, Radius server, and Billing server to complete user authentication, traffic management, and Billing.
3) for the connection between the company's branches and the company's headquarters, you can use an ATM switch on the ATM Backbone Network and the aggregation layer to create a Modem to configure IPoA and NAT, and pay for it in the form of a leased line monthly package) connect to the company's headquarters through PVC, so that you can access the Headquarters server.
The above describes the application of ATM technology in LAN and Internet.
Vi. Summary
ATM is an excellent technology for transmission, exchange, reuse, and cross-connection. At present, the ATM technology is becoming more practical in a more pragmatic manner, and its understanding of the ATM technology should also be updated in mind. The sustainable development of the Internet requires the support of ATM to improve service quality and expand capacity. ATM also needs the Internet to develop, apply, and present itself. With the increasing use of ATM, the advantages of ATM are becoming increasingly prominent. The development of the Internet is an opportunity, especially the multi-media demand of Internet businesses. It will be a huge boost to the ATM technology.