CN101409689B - Method for exchanging internet address - Google Patents

Abstract

The Internet address exchange method belongs to the field of computer network technology, and is characterized in that: based on the existing Internet protocol, Internet users are divided into a large number of ordinary users and several groups of high-performance users; ordinary users use the current "best effort" Internet The service mode remains unchanged, high-performance users configure dedicated high-performance addresses when needed, and use "high-quality" services. The method can provide different services for different users through admission control, dynamic allocation, routing release and bandwidth reservation of high-performance addresses in autonomous domains, link selection and bandwidth reservation between domains, and hourly charging. The invention effectively solves major problems such as the quality of service (QoS) problem of the current Internet, the network security problem and the sustainable development economic model, and has good scalability and stability. It provides new architecture support for the development of other important technical fields such as wireless technology and multicast technology in the Internet.

Description

internet address exchange method

technical field

A method for exchanging Internet addresses belongs to the technical field of computer networks.

Background technique

The Internet based on packet switching has now become the most important information infrastructure in the world. It has become a consensus that the three-network integration of communication network, radio and television network and computer network takes the Internet as the technical basis. The Internet has the characteristics of good scalability, survivability, flexibility, and ubiquity. However, the current Internet design concept of "no connection, end-to-end, and best effort" also brings poor security and cannot guarantee service quality. (QoS) and no clear sustainable economic model. Video, real-time applications, P2P and other applications have increasing demand for Internet bandwidth, but the improvement of router performance is limited by Moore's Law, which leads to the contradiction between the demand for unlimited bandwidth and limited bandwidth resources. At the same time, the demand for security and reliability makes this The contradictions became more intensified.

Although people have conducted many years of research in the fields of the Internet, communications, and radio and television, and have proposed many solutions to independent problems, some solutions essentially violate the design concept of the Internet and bring about scalability. And other issues. At the same time, solutions to individual problems come at the cost of making other problems worse. At present, some major projects have been launched internationally (especially in the United States), with the purpose of redesigning the Internet architecture, but in view of the scale of the Internet in the world, it is impossible to abandon the current Internet.

The Internet address exchange method proposed by the present invention fully utilizes the advantages of packet switching and circuit switching, and on the basis of not changing the existing Internet core protocol, the Internet system structure has been greatly improved, and the current Internet quality of service (QoS) is systematically solved. ) issues, security control issues and sustainable development economic model issues.

The basic features of the present invention include: 1) the Internet address exchange method is based on the current Internet protocol, and defines the parameters of high-performance network application service requirements, including the network (service domain) where the opposite end communication object of the user is located, the required bandwidth, Corresponding business volume and required service quality, such as delay, packet loss rate, jitter and other indicators. Define application services with service quality and security requirements as high-performance application services, addresses used for high-performance services as high-performance addresses, and users who apply for high-performance application services as high-performance users. 2) Internet users are divided into a large number of ordinary users and a limited number of high-performance users. Ordinary users use ordinary IP addresses to enjoy the current "best effort" Internet service model, and high-performance users use dedicated high-performance addresses to enjoy high-performance applications that guarantee end-to-end performance. 3) Calculate the required maximum number of concurrent communication processes according to the business volume of high-performance users and the allowable call loss rate. On this basis, according to the routing policy, calculate the required number of high-performance addresses. According to the bandwidth requirements of high-performance communication, the total bandwidth of high-performance communication is calculated; the high-performance address block and the total bandwidth of high-performance communication are the basis for bandwidth reservation. 4) Several specific high-performance address blocks are reserved and bound to specific high-performance network application service requirements (parameters). 5) The total bandwidth in extreme cases is reserved for high-performance addresses in the autonomous domain, and the corresponding total bandwidth is reserved for high-performance address blocks in cross-domain situations. 6) Between autonomous domains, select a cross-domain path; for the adjacent autonomous domains of the selected path, issue static, clustered high-performance address blocks through the inter-domain routing protocol (BGP); 7) Add three layers of soft The exchange server and the layer-3 admission control gateway router control the allocation and use of high-performance addresses. 8) The three-layer softswitch server verifies user identity, service domain and other information according to the needs of high-performance users, and at the same time performs admission control for high-performance users according to the maximum allowed concurrent number, and dynamically allocates high-performance address segments; 9) Three-layer The access control gateway router device enables the high-performance address segment to take effect, controls its bandwidth, and through the interaction with the routing protocol (iBGP or eBGP multihop, etc.) In the sub-domain; 10) In the cross-domain situation, the user uses a dynamically allocated high-performance address segment for high-performance communication. At this time, the performance and service quality of the user's inbound traffic can be fully guaranteed, and the performance and service quality of the user's outbound traffic The quality of service is guaranteed by the peer communication object through the implementation of the Internet address exchange method; 11) For the dynamically allocated high-performance addresses, the billing is based on the unit time bandwidth resource usage fee corresponding to a certain type of application service and the time product.

Contents of the invention

The purpose of the present invention is to systematically solve the current Internet quality of service (QoS) problem, security control problem and sustainable development economic model problem on the basis of not changing the existing Internet. The structure of the present invention as shown in Figure 1, comprises pre-configured Internet infrastructure, three-layer soft switch server in the address switching architecture, three-layer access control gateway router in the address switching architecture, and dynamic address user terminals .

The method described in this patent is based on existing Internet protocols and infrastructure. For ordinary users, it can provide "best effort" application services without service quality guarantee. For users with high performance requirements, it can provide application services with service quality assurance and security assurance. At the same time, the Internet service based on the present invention has a good economic model of sustainable development.

The present invention can be implemented in a single autonomous domain, and can also be implemented in the respective autonomous domains of both communication parties; the feature of the present invention implemented in a single autonomous domain is to include the following steps in sequence:

The Internet address exchange method is characterized in that the method is implemented in a single autonomous domain, or implemented in the respective autonomous domains of both communication parties according to the following steps:

Step (1): Initialization

Internet users are divided into a large number of ordinary users and a limited number of high-performance users. The high-performance users refer to users with high-performance application requirements such as service quality and security. Define the parameters of the high-performance application requirements , including the network service domain where the user's peer communication object is located, the required bandwidth, the corresponding business volume, and the service quality indicators including the required delay, packet loss rate, and jitter;

The address used for this high-performance application requirement is called a high-performance address, and several specific high-performance address blocks are reserved and bound to specific high-performance application requirement parameters;

Configure at least one layer-3 softswitch server in each autonomous domain, configure layer-3 admission control gateway routers in the subnet where the high-performance users are located; configure enhanced dynamic address user terminals for users who intend to apply for high-performance applications ;

Preset in the address database on the layer-3 softswitch server:

The number of high-performance addresses required by all high-performance users, which is calculated based on the business volume of high-performance users and the allowable call loss rate, blocking rate, and routing policy;

The total bandwidth of high-performance applications required by all high-performance users is obtained based on the number of concurrent high-performance users and the bandwidth requirements of single high-performance network communication;

In the link database on the three-layer softswitch server, the cross-domain route for the high-performance application routing selection is preset, and bandwidth is reserved: for the selected cross-domain link, the corresponding high-performance address is statically , Clustered BGP route announcement;

Configure routing protocols on at least one router in the autonomous domain so as to establish neighbor relationships with all Layer 3 admission control gateway routers in the autonomous domain, so that when high-performance addresses are dynamically assigned to the Layer 3 admission control gateway routers, according to The routing protocol publishes the route corresponding to the high-performance address to all routers in the autonomous domain;

The signaling transmission control module includes: a network-network signaling receiving or sending module and a user-network signaling receiving or sending module, which initializes the signaling transmission control module and waits for scheduling control of transmission signaling. The signaling format includes:

User-network signaling includes: user identification, serial number, common address of the user's local end, common address of the communication peer end, high-performance address segment of the user's local end, bandwidth required by the user's current application, and the waiting time and transmission time of the user's abnormal exit instruction;

Network - Network signaling includes: user identification, serial number, high-performance address segment of the user's local end, maximum bandwidth, and abnormal waiting time for the user to exit and transmission instructions;

Step (2): The high-performance user applies for the use of the high-performance address in turn according to the following steps:

Step (2.1): A high-performance user a in the autonomous domain X sends a user-network signaling UNI to the layer-3 softswitch server through the enhanced dynamic address user terminal, and submits an application to communicate with the high-performance user in the autonomous domain Z h for high-performance communication;

Step (2.2): after the three-layer softswitch server verifies that the identity of user a is correct, and the number of users on the link does not reach the upper limit set at this moment, select a high-performance address segment to be dynamically assigned to by the address database User a, at the same time, transmits parameters including high-performance network address segment, maximum bandwidth, and waiting time for abnormal exit to the layer-3 access control gateway of the subnet where user a is located through network-network signaling NNI Router; if the condition is not met, reject the request of user a;

Step (2.3): the three-layer admission control gateway router configures the high-performance address obtained from the three-layer softswitch server to dynamically distribute to the equipment interface connected to the subnet where the high-performance user is located, and press the maximum allowable Limit the traffic of the interface by bandwidth; through the pre-configured neighbor relationship established with the routing protocol of this autonomous domain, publish the route of the dynamically allocated high-performance address segment to each management domain on the path, and start the non- Exit timing normally;

Step (2.4): The three-layer admission control gateway router notifies the three-layer softswitch server that the configuration is complete, and the three-layer softswitch server prepares to receive and charge user a through the charging database; the three-layer softswitch server starts abnormally Exit timing;

Step (3): The high-performance user a ends using the high-performance application service in turn according to the following steps:

Step (3.1): the high-performance user a notifies the three-layer softswitch server that the high-performance application service ends through the user-network signaling UNI;

Step (3.2): the three-layer softswitch server reclaims the corresponding high-performance address, waits to be allocated to the next requesting user, and simultaneously notifies the corresponding three-layer access control gateway router to modify the configuration through the network-network signaling NNI ;

Step (3.3): the three-layer admission control gateway router terminates the network interface configuration corresponding to the high-performance address, flow control configuration and route distribution;

Step (3.4): the three-layer softswitch server stops billing, notifies the corresponding high-performance user of the billing result based on the bandwidth resource usage fee per unit time corresponding to a certain class of application services and the time product, and updates the user database, address database and meter Fee database, this high-performance address segment is released, waiting to be allocated to the next requesting user.

In the step (1), the preferred option for configuring the routing protocol for at least one router in the autonomous domain is the internal border gateway protocol iBGP, or the external border routing protocol eBGP, or the internal routing protocol IGP.

In the step (1), the total bandwidth of the high-performance communication is W>B×N, where N is the number of high-performance users, and B is the bandwidth requirement of a single high-performance network communication.

In the step (1), the maximum bandwidth that can be used by a single stream of the common address is smaller than the bandwidth provided by the high-performance address.

The experimental test results have reached the expected requirements of the present invention. By distinguishing between ordinary users and high-performance users, reserving bandwidth for high-performance addresses, performing admission control for high-performance users according to applications, dynamically assigning high-performance addresses and charging accordingly, and successfully solving the problem of high-performance users in a systematic and engineering way Applications for performance users provide quality of service (QoS) and security guarantees, and can have a good economic model.

The composition of the modules of this patent and the relationship between the modules are shown in Figure 2. The block diagram of the description method of this patent specification is shown in Figure 3. The experimental test environment is shown in Figure 4.

Description of drawings

Figure 1 is a structural diagram of the Internet address exchange network.

Figure 2 The composition of the modules of the Internet address exchange architecture and the relationship between the modules.

Figure 3 is a block diagram of the work flow of the Internet address exchange network system.

Figure 4 is a network connection topology diagram of the experimental test environment.

Fig. 5 The experimental data of user a using common address communication test.

Fig. 6 The experimental data of user a using the high-performance address communication test.

Fig. 7 The experimental data of user b using the high-performance address communication test.

Specific implementation process

The method of the present invention comprises 4 building blocks organically formed, namely " pre-configured Internet infrastructure ", " three-layer soft switch server ", " three-layer admission control gateway router " and " dynamic address user terminal ", describe respectively as follows .

(1) Pre-configured Internet infrastructure

The Internet address exchange architecture is based on the current Internet protocols and standards, with the autonomous domain as the basic implementation unit. In the autonomous domain, ordinary users and high-performance users are distinguished according to requirements, and ordinary addresses are allocated to ordinary users. In addition to assigning ordinary addresses to high-performance users, high-performance addresses can also be assigned. These high-performance addresses clearly define the network where the two communicating parties are located and the bandwidth required by specific applications, and the quality of service such as delay, packet loss rate, and jitter parameter. These high-performance addresses are reserved in the network and correspond to required services, but are not bound to specific users.

计算出所需的最大并发通信进程数量，其中a为业务量，N为并发用户数，E为阻塞率。在此基础上，根据路由政策，计算出所需的高性能地址数量。其计算公式为：P＝min(PIR，PAU-ceil(log₂ N))，其中P为所需掩码长度，PIR为跨域最大允许公布的掩码长度，PAU在IPv4的情况下为30，min()，ceil()和log₂()分别为取最小值、取下一个整数值和以2为底的对数的标准函数。根据上述并发用户数N和单个高性能网络通信的带宽需求B，计算所需跨域带宽W，即选择B×N≤W。According to the business volume of high-performance users and the allowable call loss rate, according to the Irish formula

Calculate the maximum number of concurrent communication processes required, where a is the business volume, N is the number of concurrent users, and E is the blocking rate. On this basis, according to the routing policy, calculate the required number of high-performance addresses. The calculation formula is: P=min(PIR, PAU-ceil(log ₂ N)), where P is the required mask length, PIR is the maximum allowable cross-domain mask length, and PAU is 30 in the case of IPv4 , min(), ceil(), and _log2 () are the standard functions for taking the minimum value, taking the next integer value, and taking the logarithm to base 2, respectively. According to the above-mentioned number of concurrent users N and the bandwidth requirement B of a single high-performance network communication, calculate the required cross-domain bandwidth W, that is, choose B×N≤W.

In the autonomous domain, use the service class function (Service Class) of the router to reserve sufficient bandwidth for high-performance address blocks in extreme cases, that is, no matter which high-performance user the high-performance address is assigned to, the required bandwidth can be guaranteed.

Select one or several routers for each management domain in the autonomous domain and configure the routing distribution with the Layer 3 access control gateway router device, so as to dynamically publish the routing information of high-performance addresses. According to the situation, the internal border routing protocol (iBGP) is preferred. The route reflection server is used as the route distribution mechanism.

Between autonomous domains, select a specific link, and use the service class function (ServiceClass) of the router on this link to reserve sufficient bandwidth for high-performance address blocks, and perform static and clustered high-speed Performance Block Boundary Routing Protocol (BGP) route announcement.

Use the current standard quality of service configuration method and bandwidth reservation method on the router, or the Internet quasi-minimum state flow control method, to control the flow of ordinary addresses within or between domains.

(2) Layer-3 softswitch server in the address exchange method

The three-layer softswitch server in the address switching architecture is a dedicated computer server system. Each autonomous domain is configured with one or more networked layer-3 softswitch servers. It consists of a signaling transmission scheduling module, a user identity authentication module, a domain and cross-domain link bandwidth usage statistics module, a billing module, a domain management module and a control module. In addition to the signaling transmission scheduling module, the core of each functional module is a database system, including: user database, address database, link database, access control gateway database and billing database. The signaling system includes: user-network signaling system (UNI) and network-network signaling system (NNI).

The user database stores user accounts and authentication information, and establishes a relationship with corresponding billing information. The address database stores high-performance addresses and establishes relationships with links and the concurrency supported by the addresses. The link database stores the information of the cross-domain link with reserved bandwidth in the autonomous domain and the common address range of the local end and the opposite end using this link, and establishes a relationship with the address data. The layer-3 access control gateway router database stores the information of the layer-3 access control gateway router equipment in the subnet where each high-performance user in the autonomous domain resides, and establishes a relationship with user data. The billing database establishes relationships with user data. The three-layer soft switch server accepts the request of registering the user terminal with the dynamic address of the high-performance user through the user-network signaling (UNI). Bandwidth required for performance communication, waiting time for abnormal exit, etc. An example of the format described in the XML language is:

<uni>

<customer id="customerid">//user ID

</customer>

<order id="ordered">//serial number

</order>

<metric name=″own-address″>//Ordinary address of the user’s local end

x.x.x.x

</metric>

<metric name="end-address">//Ordinary address of the peer end of the user

y.y.y.y

</metric>

<metric name=″hp-prefix″>//user-side high-performance address segment

x.x.x.x/k

</metric>

<metric name=″bandwidth″units=″Mbps″>//User demand bandwidth

30

</metric>

<metric name=″ttl″units=″second″>//The waiting time for the user to exit abnormally

3600

</metric>

<metric name="command">//command

confirm

</metric>

</uni>

The general user database, link database, and layer-3 access control gateway router database of the layer-3 softswitch server verify the user's identity, verify the range of the user's local common address and peer-end common address to determine whether there is a layer-3 access control gateway router device and For the corresponding link, determine whether high-performance addresses can be allocated according to the maximum concurrent number of the corresponding address and the current concurrent number. If the verification is passed, the layer-3 softswitch server controls the corresponding layer-3 access control gateway router device through network-network signaling (NNI), and its parameters are: high-performance address segment, maximum bandwidth, and waiting time for abnormal exit wait. An example of the format described in its XML language is:

<nni>

<customer id="customerid">//user ID

</customer>

<order id="ordered">//serial number

</order>

<metric name=″prefix″unit=″IPv4″>//The user's local high-performance address segment

z.z.z.z/m

</metric>

<metric name=″max_bandwidth″unit=″Mbps″>//maximum bandwidth

30

</metric>

<metric name=″ttl″units=″second″>//User abnormal logout waiting time

3600

</metric>

<metric name="command">//command

active

</metric>

</nni>

(3) Layer-3 admission control gateway router in the address exchange method

The L3 admission control gateway router in the address switching architecture is a special router. The routing function on the existing router is implemented by loading modules related to signaling transmission processing. It consists of four parts: control module, interface module, routing module and signaling transmission scheduling module. The control module includes: user abnormal exit timer and network management client module; the interface module includes: network interface address configuration and flow control module; the routing module includes: routing protocol and routing distribution module; the signaling transmission scheduling module is: network-network Signaling system (NNI) and user-network signaling system.

Layer-3 admission control gateway routers are set in each high-performance user's subnet. According to statistics, the traffic of Internet users is distributed according to a power law, that is, ordinary users are the vast majority, and high-performance users are a small number. Therefore, the total number of Layer 3 access control gateway routers is limited.

The Layer 3 admission control gateway router establishes a neighbor relationship with the route reflection server of the internal border routing protocol (iBGP) of the administrative domain on all possible paths in the autonomous domain, so the Layer 3 admission control gateway router can forward the information on the device to the network Routes are advertised to the management domain on all possible paths in the autonomous domain.

The subnet where each registered high-performance user is located is equipped with a networked layer-3 admission control gateway router, which communicates with the layer-3 softswitch server that manages it through network signaling (NNI), and allocates and revokes high-level traffic under its control. Performance address, publishing or canceling related routes, performing or revoking port flow control. An example of the format described in the XML language is:

<nni>

<customer id="customerid">//user ID

</customer>

<order id="ordered">//serial number

</order>

<metric name=″prefix″unit=″IPv4″>//The user's local high-performance address segment

z.z.z.z/m

</metric>

<metric name=″max_bandwidth″unit=″Mbps″>//maximum bandwidth

30

</metric>

<metric name=″ttl″units=″second″>//User abnormal logout waiting time

3600

</metric>

<metric name="command">//command

active

</metric>

</nni>

(4) Dynamic address user terminal

The dynamic address user terminal in the address switching architecture is an extended common user terminal device. The dynamic address user terminal equipment not only includes all the functions of the current common user terminal, but also includes address selection and configuration modules, user-network interface (UNI) signaling modules, etc.; where the UNI communicates with the softswitch server or the three-layer access control gateway router The communication, address selection and configuration modules determine the address used by the application at that time. An example of the format described in the XML language is:

<uni>

<customer id="customerid">//user ID

</customer>

<order id="ordered">//serial number

</order>

<metric name=″own-address″>//Ordinary address of the user’s local end

x.x.x.x

</metric>

<metric name="end-address">//Ordinary address of the peer end of the user

y.y.y.y

</metric>

<metric name=″hp-prefix″>//user-side high-performance address segment

x.x.x.x/k

</metric>

<metric name="bandwidth"units="Mbps">//user demand bandwidth

30

</metric>

<metric name=″ttl″units=″second″>//User abnormal logout waiting time

3600

</metric>

<metric name="command">//command

apply

</metric>

</uni>

A user terminal with a dynamic address can be configured with multiple network addresses, including common addresses through static configuration or dynamic configuration (based on state such as DHCP or stateless) and high-performance addresses. The dynamic address user terminal has a standard user-network signaling (UNI) module, which can communicate with the three-layer soft switch server in the address switching architecture, and apply for or cancel the use of high-performance addresses. The dynamic address user terminal has a dynamic address configuration module driven by signaling, so that the address is bound to the application, that is, ordinary applications use ordinary addresses, and high-performance applications use high-performance addresses.

The test environment is set up on the real Internet backbone network. The local users are high-performance users in two different locations of CERNET in Beijing, and the opposite end is the high-performance video server of the TEIN2 Singapore node. Video application uses DVTS system, 1 channel video is 30Mbps. The test environment is shown in Figure 4.

Due to the limitation of network bandwidth, the maximum available bandwidth of CERNET ordinary users through TEIN2 is 10Mbps, so they cannot watch video images transmitted from Singapore. In this experiment, the domain of the three-layer softswitch server in the address switching architecture is selected as CERNET, and the high-performance application concurrent process is 1. There are 2 Layer 3 admission control gateway routers in total. According to the BGP routing policy PIR=24 of CERNET and TEIN2. Using IPv4, the high-performance address segment selected by PAU=30 is 202.38.112.64/30. Note that the static, clustered address advertised to TEIN2 is 202.38.112.0/24. The subnet gateways of the two high-performance users in Beijing are 202.112.35.34 and 202.38.97.253 respectively. There are three cases in this test:

1) User a of the high-performance application in Beijing uses a common address to receive DVTS video from the Singapore node, as shown in Figure 5. Note that the iBGP route reflector of the autonomous domain AS4538 assigns 0 high-performance addresses to neighbors 202.38.97.253 (high-performance application user a) and 202.38.35.34 (high-performance application user b) of the Layer 3 access control gateway router. (as shown in Figure 5(a)). The total number of routing entries in the autonomous domain is 14155 (as shown in Figure 5(b)), and the total number of routing entries published by the autonomous domain AS4538 to TEIN2 is 203 (as shown in Figure 5(c)). The high-performance address segment 202.38.112.64/30 does not appear in the autonomous domain AS4538 (as shown in Figure 5(d)). Since ordinary addresses cannot support a bandwidth greater than 10Mbps, the video image quality is very poor at this time (as shown in Figure 1 in the attachment).

2) High-performance user a in Beijing applies for a high-performance address to receive DVTS video from the Singapore node, as shown in Figure 6. At this time, high-performance user a applies for a high-performance address to the layer-3 softswitch server through UNI, and the content is as follows:

<uni>

<customer id="BJ-a">//User ID

</customer>

<order id="12345">//serial number

</order>

<metric name=″own-address″>//Ordinary address of the user’s local end

202.38.97.254

</metric>

<metric name="end-address">//Ordinary address of the peer end of the user

202.179.252.102

</metric>

<metric name="bandwidth"units="Mbps">//user demand bandwidth

30

</metric>

<metric name=″ttl″units=″second″>//User abnormal logout waiting time

3600

</metric>

<metric name="command">//command

apply

</metric>

</uni>

The Layer 3 softswitch server verifies that BJ-a is a valid user through the user database, link database, Layer 3 access control gateway router database, and address database, which matches the local common address 202.38.97.254, and the corresponding subnet has been configured with three Layer access control gateway router equipment, the peer common address 202.179.252.102 has a valid link to provide high-performance applications, the maximum allowable bandwidth is 35Mbps, the corresponding high-performance address segment is 202.38.112.64/30, and the maximum allowable number of concurrent communications is 1, and the current concurrency number is 0. Therefore, the layer-3 softswitch server can allow user BJ-a to use the high-performance address 202.38112.64/30 to communicate. The layer-3 softswitch server notifies the layer-3 admission control gateway router device of the subnet where user a is located through the NNI.

<nni>

<customer id="BJ-a">//User ID

</customer>

<order id="12345">//serial number

</order>

<metric name=″prefix″unit=″IPv4″>//The user's local high-performance address segment

202.38.112.64/30

</metric>

<metric name=″max_bandwidth″unit=″Mbps″>//maximum bandwidth

35

</metric>

<metric name=″ttl″units=″second″>//User abnormal logout waiting time

3600

</metric>

<metric name="command">//command

Request

</metric>

</nni>

The layer-3 access control gateway router implements processes such as address allocation, flow control, and route distribution, and replies to the layer-3 softswitch server for confirmation.

<nni>

<customer id="BJ-a">//User ID

</customer>

<order id="12346">//serial number

</order>

<metric name="command">//command

Confirm

</metric>

</nni>

At this time, the layer-3 softswitch server notifies user a in Beijing through UNI, and starts billing.

<uni>

<customer id="BJ-a">//User ID

</customer>

<order id="12346">//serial number

</order>

<metric name=″hp-prefix″>//user-side high-performance address segment

202.38.112.64/30

</metric>

<metric name="command">//command

Confirm

</metric>

</uni>

User a can start a high-performance application. As shown in Figure 6, the iBGP route reflector of the autonomous domain AS4538 assigns 1 entry to the high-performance address of the Layer 3 admission control gateway router neighbor 202.38.97.253 (high-performance application user a), but to 202.38.35.34 ( The number of high-performance address allocations for high-performance application user b) is 0 (as shown in Figure 6(a)). The total number of routes in the autonomous domain is 14156 (one added) (as shown in Figure 6(b)), and the total number of routes advertised by the autonomous domain AS4538 to TEIN2 is 203 (not increased) (as shown in Figure 6(c)). The high-performance address segment 202.38.112.64/30 appears in the autonomous domain AS4538, and the corresponding next hop is 202.38.97.253 (as shown in Figure 6(d)). Due to the use of high-performance addresses, it can support a bandwidth greater than 30Mbps, and the video image quality is very good at this time (as shown in Figure 2 in the attachment).

At this time, user b also applies for using a high-performance address, and notifies the layer-3 softswitch server through UNI.

<uni>

<customer id="BJ-b">//User ID

</customer>

<order id="20001">//serial number

</order>

<metric name=″own-address″>//Ordinary address of the user’s local end

202.112.35.34

</metric>

<metric name="end-address">//Ordinary address of the peer end of the user

202.179.252.102

</metric>

<metric name=″hp-prefix″>//user-side high-performance address segment

0.0.0.0/0

</metric>

<metric name="bandwidth"units="Mbps">//user demand bandwidth

30

</metric>

<metric name=″ttl″units=″second″>//User abnormal logout waiting time

3600

</metric>

<metric name="command">//command

apply

</metric>

</uni>

The layer-3 softswitch server verifies that BJ-b is a valid user through the user database, link database, layer-3 access control gateway router database, and address database, which matches the local common address 202.112.25.34, and the corresponding subnet has been configured with three Layer access control gateway router equipment, the peer common address 202.179.252.102 has a valid link to provide high-performance applications, the maximum allowable bandwidth is 35Mbps, the corresponding high-performance address segment is 202.38.112.64/30, and the maximum allowable number of concurrent communications is 1, and the current concurrency number is 1. Therefore, the layer-3 softswitch server cannot allocate high-performance addresses for user b. Deny the service request of user b.

<uni>

<customer id="BJ-b">//Customer ID

</customer>

<order id="20002">//serial number

</order>

<metric name="command">//command

Refuse

</metric>

</uni>

User b has to wait for a certain amount of time to request again.

When user a completes the high-performance communication, similar to the above process, user a uses UNI to notify the layer-3 softswitch server to withdraw the high-performance address.

<uni>

<customer id="BJ-a">//User ID

</customer>

<order id="12347">//serial number

</order>

<metric name=″own-address″>//Ordinary address of the user’s local end

202.38.97.254

</metric>

<metric name="end-address">//Ordinary address of the peer end of the user

202.179.252.102

</metric>

<metric name=″hp-prefix″>//user-side high-performance address segment

202.38.112.64/30

</metric>

<metric name="command">//command

Stop

</metric>

</uni>

The layer-3 softswitch server uses the NNI to notify the corresponding layer-3 admission control gateway router device.

<nni>

<customer id="BJ-a">//User ID

</customer>

<order id="12348">//serial number

</order>

<metric name=″prefix″unit=″IPv4″>//The user's local high-performance address segment

202.38.112.64/30

</metric>

<metric name="command">//command

Stop

</metric>

</nni>

The access control gateway device revokes the high-performance address, stops publishing the high-performance address, and confirms to the layer-3 softswitch server through the NNI.

<nni>

<customer id="BJ-a">//User ID

</customer>

<order id="12349">//serial number

</order>

<metric name="command">//command

Confirm

</metric>

</nni>

The layer-3 softswitch server notifies user a with UNI to stop billing, and the fee is bandwidth × usage time.

<uni>

<customer id="BJ-a">//User ID

</customer>

<order id="12348">//serial number

</order>

<metric name="command">//command

Stop

</metric>

</uni>

3) Another high-performance user b in Beijing applies for a high-performance address to receive DVTS video from the Singapore node. At this point, high-performance user a has completed high-performance communication. Similar to the communication process between UNI and NNI in the above 2), user b can obtain a performance address for communication, as shown in FIG. 7 . At this time, the iBGP route reflector of autonomous domain AS4538 assigns 0 entries to the high-performance address of the neighbor 202.38.97.253 (high-performance application user a) of the Layer 3 admission control gateway router, but to 202.38.35.34 (high-performance application user a) The number of high-performance address allocations for user b) is 1 (as shown in Figure 7(a)). The total number of routes in the autonomous domain is 14156 (one added) (as shown in Figure 7(b)), and the total number of routes advertised by the autonomous domain AS4538 to TEIN2 is 203 (not increased) (as shown in Figure 7(c)). The high-performance address segment 202.38.112.64/30 appears in the autonomous domain AS4538, and the corresponding next hop is 202.112.35.34 (as shown in Figure 7(d)). Due to the use of high-performance addresses, it can support a bandwidth greater than 30Mbps, and the video image quality is very good at this time (as shown in Figure 3 in the attachment).

Among them, the high-performance address is allocated to user a and the change of the relevant routing table of user b clearly shows the working process of the address switching architecture, the role played by the three-layer soft switch, three-layer admission control gateway router equipment and dynamic address user terminal . It can be seen from the figure that the video image quality of the ordinary address cannot meet the requirements, but the video image quality of the high-performance address is very good.

Claims (4)

1. The Internet address exchange method is characterized in that the method is implemented in a single autonomous domain, or implemented in the respective autonomous domains of both communication parties according to the following steps: Step (1): Initialization Internet users are divided into a large number of ordinary users and a limited number of high-performance users. The high-performance users refer to users with high-performance application requirements such as service quality and security. Define the parameters of the high-performance application requirements , including the network service domain where the user's peer communication object is located, the required bandwidth, the corresponding business volume, and the service quality indicators including the required delay, packet loss rate, and jitter; The address used for this high-performance application requirement is called a high-performance address, and several specific high-performance address blocks are reserved and bound to specific high-performance application requirement parameters; Configure at least one layer-3 softswitch server in each autonomous domain, configure layer-3 admission control gateway routers in the subnet where the high-performance users are located; configure enhanced dynamic address user terminals for users who intend to apply for high-performance applications ; Preset in the address database on the layer-3 softswitch server: The number of high-performance addresses required by all high-performance users, which is calculated based on the business volume of high-performance users and the allowable call loss rate, blocking rate, and routing policy; The total bandwidth of high-performance applications required by all high-performance users is obtained based on the number of concurrent high-performance users and the bandwidth requirements of single high-performance network communication; In the link database on the three-layer softswitch server, the cross-domain route for the high-performance application routing selection is preset, and bandwidth is reserved; for the selected cross-domain link, the corresponding high-performance address is statically , Clustered BGP route announcement; Configure routing protocols on at least one router in the autonomous domain so as to establish neighbor relationships with all Layer 3 admission control gateway routers in the autonomous domain, so that when high-performance addresses are dynamically assigned to the Layer 3 admission control gateway routers, according to The routing protocol publishes the route corresponding to the high-performance address to all routers in the autonomous domain; The signaling transmission control module includes: a network-network signaling receiving or sending module and a user-network signaling receiving or sending module, which initializes the signaling transmission control module and waits for scheduling control of transmission signaling. The signaling format includes: User-network signaling includes: user identification, serial number, common address of the user's local end, common address of the communication peer end, high-performance address segment of the user's local end, bandwidth required by the user's current application, and the waiting time and transmission time of the user's abnormal exit instruction; Network-Network signaling includes: user identification, serial number, high-performance address segment of the user's local end, maximum bandwidth, abnormal waiting time for exiting the user, and transmission instructions: Step (2): The high-performance user applies for the use of the high-performance address in turn according to the following steps: Step (2.1): A high-performance user a in the autonomous domain X sends a user-network signaling UNI to the layer-3 softswitch server through the enhanced dynamic address user terminal, and submits an application to communicate with the high-performance user in the autonomous domain Z h for high-performance communication; Step (2.2): after the three-layer softswitch server verifies that the identity of user a is correct, and the number of users on this link has not reached the upper limit set, select a high-performance address segment to be dynamically assigned to by the address database User a, at the same time, transmits parameters including high-performance network address segment, maximum bandwidth, and waiting time for abnormal exit to the layer-3 access control gateway of the subnet where user a is located through network-network signaling NNI Router; if the condition is not met, reject the request of user a; Step (23): the three-layer admission control gateway router configures the high-performance address obtained from the three-layer softswitch server to dynamically distribute to the device interface connected to the subnet where the high-performance user is located, and press the maximum allowable Limit the traffic of the interface by bandwidth; through the pre-configured neighbor relationship established with the routing protocol of this autonomous domain, publish the route of the dynamically allocated high-performance address segment to each management domain on the path, and start the non- Exit timing normally; Step (2.4): The three-layer admission control gateway router notifies the three-layer softswitch server that the configuration is complete, and the three-layer softswitch server prepares to receive and charge user a through the charging database; the three-layer softswitch server starts abnormally Exit timing; Step (3): The high-performance user a ends using the high-performance application service in turn according to the following steps: Step (3.1): the high-performance user a notifies the three-layer softswitch server that the high-performance application service ends through the user-network signaling UNI; Step (3.2): the three-layer softswitch server reclaims the corresponding high-performance address, waits to be allocated to the next requesting user, and simultaneously notifies the corresponding three-layer access control gateway router to modify the configuration through the network-network signaling NNI : Step (3.3): the three-layer admission control gateway router terminates the network interface configuration corresponding to the high-performance address, flow control configuration and route distribution; Step (3.4): the three-layer softswitch server stops billing, notifies the corresponding high-performance user of the billing result based on the bandwidth resource usage fee per unit time corresponding to a certain class of application services and the time product, and updates the user database, address database and meter Fee database, this high-performance address segment is released, waiting to be allocated to the next requesting user. 2. Internet address exchange method according to claim 1, it is characterized in that, in said step (1) for at least one router configuration routing protocol preference in this autonomous domain is internal border gateway protocol iBGP, or external boundary The routing protocol eBGP, or the interior routing protocol IGP. 3. Internet address exchange method according to claim 1, is characterized in that, in described step (1), described high-performance communication total bandwidth W＞B*N, N is described high-performance user number, B Bandwidth requirements for communication over a single high-performance network. 4. The Internet address exchange method according to claim 1, characterized in that, in said step (1), the maximum bandwidth that can be used by a single stream of common addresses is less than the bandwidth provided by high-performance addresses.

Images