CN106685571A - A hybrid protection method based on recovery time sensitivity in WDM optical network - Google Patents

Abstract

The invention discloses a hybrid protection method in a WDM (wavelength division multiplexing) optical network based on recovery time sensitivity, and belongs to the technical field of communication networks. The method is based on the recovery time constraint of business connection, aims at the conflict between the recovery time and the resource utilization rate, employs a mode of combining shared access protection and dedicated access protection, provides a flexible backup (protection) resource configuration scheme for businesses at different levels, optimizes the network resource utilization rate under the condition that a business recovery time threshold is met, and provides a hybrid protection configuration, which discriminates the recovery degrees, for the business connection.

Description

A hybrid protection method based on recovery time sensitivity in WDM optical network

technical field

The invention relates to the technical field of communication networks, in particular to a hybrid protection method based on recovery time sensitivity in a WDM optical network.

Background technique

In the optical network, which is the infrastructure of the telecommunications backbone network, once a component fails or is damaged, it will cause a large amount of data loss and service interruption. Therefore, the survivability problem has always been an important research topic in WDM optical networks. Network survivability refers to the ability to maintain the necessary quality of service when a network failure occurs. In the study of network survivability, protection technology, as an important invulnerability technology, has received extensive research and attention. Protection refers to reserving backup (protection) resources for service connections in the network before a failure occurs. When a network failure occurs, the damaged service is switched to the reserved protection resource to realize service recovery.

There are two main methods of path protection: dedicated path protection (DPP) and shared path protection (SPP). Dedicated path protection (DPP) means that the reserved wavelength resources on the protection path and the OXC (optical cross-connect) on the node have been configured and are dedicated and exclusively shared by one service connection. As shown in Figure 1, the path a-b-c-d is the working path connecting the node pair (a, d), and the path a-e-f-g-d is the corresponding dedicated protection path. When a link b-c on the working path fails, node b will send alarm message, and the source node a initiates the switching of the service path. Since the OXC on the nodes of the backup path in the dedicated protection has been configured in advance, the business on the working path will be switched to the backup path immediately, so the service recovery time is faster but the backup resource consumption is large. Shared path protection (SPP) means that the protection paths of some services can share the reserved wavelength resources in the network under certain conditions, but the OXC needs to be configured before the service switching. As shown in Figure 2, the path a-b-c-d is the working path connecting the node pair (a, d), a-e-g-d is the corresponding backup path, the path h-o-p-q is the working path connecting the node pair (h, q), and h-e-g-q is the corresponding backup path. Since the two paths a-b-c-d and h-o-p-q are separated, the backup resource on link l can be shared by connection (a,d) and connection (h,q) under single link failure. When a link b-c on the working path a-b-c-d fails, node b will send an alarm message to source node a, and source node a will trigger the configuration of OXC on the backup path, and then switch the business path, so the backup resources for protection are shared Less consumption but longer service recovery time.

Due to the exclusive protection of backup resources, service connection configuration consumes a lot of network resources, especially when network resources are scarce, a large number of services will not be able to establish connections due to lack of sufficient resources. The shared protection has the characteristics of service switching time lag, and the temporary interruption time of the service will be extended to varying degrees, especially for high-level service connections, it will be difficult to guarantee the corresponding SLA (service level agreement). The occurrence of any of the above situations will cause great losses to users and operators.

In order to solve the contradiction between traditional dedicated protection and shared protection in business recovery time and resource utilization, we propose a hybrid protection method based on recovery time sensitivity. Based on the recovery time constraints of service connections, this method adopts the combination of shared channel protection and dedicated channel protection to provide flexible backup resource configuration schemes for different levels of services, and optimize the use of network resources on the premise of meeting the service recovery time threshold. rate, providing service connections with a hybrid protection configuration that distinguishes recovery degrees.

Contents of the invention

The present invention aims to solve the above problems in the prior art, and proposes a hybrid protection method aiming at multi-link failure, controlling recovery time, improving network resource utilization, and distinguishing service recovery degrees. Technical scheme of the present invention is as follows:

A WDM (wavelength division multiplexing) optical network based on recovery time-sensitive hybrid protection method, which comprises the following steps:

101. Use the Dijkstra algorithm in the optical network G(N,L) to find three separate paths for each service connection c _i , namely: the working path WP _i , the first backup path BP _i ¹ and the second backup path BP _i ² , and BP _i ² is not shorter than BP _i ¹ , 1≤i≤|C|, C represents the set of all business connections in the network, N represents the set of all nodes in the network, L represents all bidirectional links in the network gather;

102. Comparing the mutual positional relationship between all service connections and arbitrary paths in the optical network G(N,L) (including: working path WP _i , first backup path BP _i ¹ and second backup path BP _i ² ), and obtaining each service Connect the shareable relationship of backup resources, and calculate the maximum recovery time RT _i ¹ and RT _i ² corresponding to the first backup path BP _i ¹ and the second backup path BP _i ² in each business connection;

103. According to the sharable relationship of the first and second backup paths, they are shared or dedicatedly configured in combination with the service recovery time threshold;

104. On any link l in the optical network G(N,L), allocate service bandwidth for all working paths WP _i passing through the link is the original service bandwidth requirement, l∈L, and adopts the resource optimization ILP model to calculate and obtain the optimal allocated bandwidth on the corresponding first backup path BP _i ¹ and the second backup path BP _i ² Among them, k is the service level, β _k is the backup channel bandwidth allocation ratio, and 0≤β ₁ ≤β ₂ ≤β ₃ ...≤β _k ≤1;

105. On any link l in the optical network G(N,L), allocate shared backup bandwidth resources for related service connections, satisfying and Among them, l∈L, k and k' represent different service levels, β _k' and β _k' represent the backup path bandwidth allocation ratios of different levels of services, and are the backup bandwidths of connections _ci and _ci' passing through link l respectively.

106. On any link l in the optical network G(N,L), allocate dedicated backup bandwidth resources for relevant business connections, satisfying Among them, l∈L, k and k' represent different service levels, β _k' and β _k' represent the backup path bandwidth allocation ratios of different levels of services, and are the backup bandwidths of connections _ci and _ci' passing through link l respectively.

Further, step 102 judges whether the backup resource can be shared according to the positional relationship of the path:

1) Rule one

Positional relationship:

or:

or

Sharing relationship:

2) Rule two

Positional relationship:

Sharing relationship:

3) Rule three

Positional relationship:

or:

or or

Sharing relationship:

4) Rule Four

Positional relationship:

Sharing relationship:

5) Rule five

Positional relationship: any other positional relationship other than the above positional relationship.

Sharing relationship:

Further, in step 103, according to the sharable relationship of the first and second backup paths, combined with the service recovery time threshold, the shared or dedicated configuration settings are performed, specifically including:

For any service connection c _i , if the second backup path BP _i ² can be shared, and RT _i ² ≤ RT ₀ , and RT ₀ is the service recovery time threshold, then set BP _i ² as shared configuration; otherwise, set BP _i ² is a dedicated configuration; if BP _i ¹ can be shared and RT _i ¹ ≤ RT ₀ , then set BP _i ¹ as a shared configuration; otherwise, set BP _i ¹ as a dedicated configuration.

Advantage of the present invention and beneficial effect are as follows:

In the present invention, aiming at the contradiction between recovery time and resource utilization, based on the recovery time constraints of service connections, a combination of shared path protection and dedicated path protection is adopted to provide flexible backup resource configuration schemes for different levels of business. Under the premise of meeting the service recovery time threshold, optimize the utilization of network resources and provide service connections with a hybrid protection configuration that distinguishes recovery degrees.

Description of drawings

Fig. 1 is a schematic diagram of dedicated path protection in the prior art;

FIG. 2 is a schematic diagram of shared path protection in the prior art;

Fig. 3 is a schematic diagram of the service recovery process of the preferred embodiment provided by the present invention;

Fig. 4 is a flow chart of a hybrid protection method based on recovery time sensitivity in a WDM optical network according to a preferred embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention will be described clearly and in detail below with reference to the drawings in the embodiments of the present invention. The described embodiments are only some of the embodiments of the invention.

The technical scheme that the present invention solves the problems of the technologies described above is:

A non-limiting embodiment is given below in conjunction with the accompanying drawings to further illustrate the present invention. Where k=3, that is, 0≤β ₁ ≤β ₂ ≤β ₃ ≤1, RT ₀ =50ms.

The concepts and models involved in the content of the present invention are as follows:

1. Damaged business recovery process

As shown in Figure 3, assume a connection from a to d, where a is the source node and d is the destination node. The recovery process is divided into two steps: the first step, if the link b-c fails on the working path, the end nodes b and c will first detect the failure, and send the failure message to the source node a and the destination node d respectively . In the second step, when the source node a receives the fault alarm message, it triggers the protection switching process of the service.

The protection switching process of the service is divided into two situations: For dedicated protection, after receiving the fault alarm message, the source node s sends a notification message to the destination node d along the backup path a-e-f-g-d. After the destination node d receives the message, the original path returns a confirmation message to the source node a, and then the source node a starts the path switching, thereby completing the restoration of the damaged service. For shared protection, after the source node a receives the fault alarm message, it sends a configuration message to the destination node d along the backup path a-e-f-g-d, and configures each OXC on the backup path step by step. When the destination node d completes the corresponding configuration, the original path A confirmation message is returned to the source node a, and then the source node a initiates path switching, thereby completing the restoration of the damaged service.

2. Network model

Assume that the WDM optical network model is G(N,L), where N is the set of nodes in the network, and L is the set of bidirectional links in the network. Each node in the network has full wavelength conversion capability, and each link has the same wavelength bandwidth. Network failures are represented by single-link failures and double-link failures.

In G(N,L), use the Dijkstra algorithm to find three separate paths for each business connection c _i (1≤i≤|C|), namely: the working path WP _i , the first backup path BP _i ¹ and The second backup path BP _i ² , and BP _i ² is not shorter than BP _i ¹ .

3. Recovery time model

Recovery time refers to the time it takes to switch the damaged service in the network to the backup (protection) path when the network fails. The relevant symbols are defined as follows:

FT: time to detect a fault;

DT: On a node, the signal processing time;

PT: the time of signal propagation on a link;

CT: the time of OXC configuration;

m: the number of hops on the working path;

n: the number of hops on the backup path;

The maximum service recovery time of connection _ci .

The calculation of the maximum service recovery time of dedicated protection is shown in formula (1), and the calculation of maximum service recovery time of shared protection is shown in formula (2).

Technical scheme of the present invention is described as follows:

The invention is a recovery time-sensitive protection scheme applied to the WDM optical network.

1. Protection channel sharing scheme:

Explanation of related symbols:

l: a bidirectional optical fiber link in the network G(N,L);

WP _i or WP _i′ : the working path of business connection i and i′;

BP _i ¹ or The first backup path corresponding to WP _i or WP _i' , that is, after WP _i or WP _i' fails, the service will be preferentially switched to the backup path;

BP _i ² or The second backup path corresponding to WP _i or WP _i' , that is, WP _i or WP _i' fails, and BP _i ¹ or Also fails, the business will switch to the backup channel;

Indicates that there are one or more links overlapping between path A and path B;

Indicates that there is no link overlap between path A and path B;

(A,B)=0: indicates that path A and path B cannot share backup resources;

(A,B)=1: indicates that path A and path B can share backup resources;

RT _i or RT _i' : the maximum recovery time of service connection i or i';

RT ₀ : service recovery time threshold, for example: 50ms.

Whether the protection path of the service connection in the present invention can adopt the resource sharing configuration scheme will depend on two conditions, that is: first judge whether the backup resource can be shared according to the positional relationship of each path; then under the sharing relationship, calculate the maximum recovery of the service Whether the time is less than the set threshold is used to determine whether the shared configuration can be adopted.

Determine whether backup resources can be shared according to the location relationship of the path:

6) Rule 1

Positional relationship:

or:

or or

Sharing relationship:

7) Rule Two

Positional relationship:

Sharing relationship:

8) Rule Three

Positional relationship:

or:

or or

Sharing relationship:

9) Rule Four

Positional relationship:

Sharing relationship:

10) Rule five

Positional relationship: any other positional relationship other than the above positional relationship.

Sharing relationship:

According to the above sharing rules, obtain the sharable relationship between each business channel, and calculate the maximum recovery time of each business connection under this relationship, and then judge whether the corresponding sharing configuration can be adopted:

Step 1: If BP _i ² can be shared, calculate the corresponding maximum recovery time RT _i ² ;

Step 2: If RT _i ² ≤ RT ₀ , then BP _i ² can adopt shared configuration, otherwise adopt dedicated configuration;

Step 3: If BP _i ¹ can be shared, calculate the corresponding maximum recovery time RT _i ¹ ;

Step 4: If RT _i ¹ ≤ RT ₀ , then BP _i ¹ can adopt shared configuration, otherwise adopt dedicated configuration;

2. Bandwidth resource allocation scheme

In order to realize the optimal allocation of bandwidth resources, the present invention uses an ILP optimization model to allocate corresponding working bandwidth and backup bandwidth to all service connections in the network. The specific model design is as follows:

enter:

G(N,L): N={n} is a node set, L={l} is a two-way link set;

C: service connection request set {c _i }(1≤i≤|C|);

· The set of all connections that the backup path passes through the link l;

W: the bandwidth capacity of each link, that is, the total number of wavelengths;

· The working bandwidth requirement of connection c _i ;

· _k : the service level of the connection ci. Set k={1,2,3} in this embodiment, wherein, 1 represents common service, 2 represents advanced service, and 3 represents emergency service;

· Indicates that the backup path _j connected to ci and the backup path q connected to _ci′ share bandwidth resources, otherwise, they do not share;

output:

· The working bandwidth allocated by connection _ci on link l;

· Backup bandwidth allocated by connection _ci on link l;

β _k : backup bandwidth allocation ratio of k-class services;

Objective function:

In formula (3), by maximizing β ₁ , β ₂ , β ₃ , various types of services can obtain more allocation of protection resources, and choosing appropriate weight coefficients (such as: 100, 70, 50) can be used in limited In the case of limited network resources, more protection resources are given priority to high-level services, and higher service recovery guarantees are provided. By _limiting the consumption of the backup resources B1 of each link to a certain extent, the sharing of backup resources is promoted, so as to realize the optimization of overall network resources.

Restrictions:

Equation (4) ensures that the working bandwidth allocated by connection _ci meets the original bandwidth requirement.

0≤β ₁ ≤β ₂ ≤β ₃ ≤1 (5)

Equation (5) guarantees that high-level services get more backup resources preferentially.

q=1or 2, j=1or 2

Equation (6) gives the backup resource sharing constraints.

Equation (7) limits the amount of bandwidth resources allocated on link l to not exceed the link capacity.

A WDM optical network based on recovery time-sensitive hybrid protection method, its specific implementation method comprises the following steps:

101. Use the Dijkstra algorithm in the optical network G(N,L) to find three separate paths for each service connection c _i , namely: the working path WP _i , the first backup path BP _i ¹ and the second backup path BP _i ² , and BP _i ² is not shorter than BP _i ¹ , 1≤i≤|C|, C represents the set of all business connections in the network, N represents the set of all nodes in the network, L represents all bidirectional links in the network gather;

102. Comparing the mutual positional relationship between all service connections and arbitrary paths in the optical network G(N,L) (including: working path WP _i , first backup path BP _i ¹ and second backup path BP _i ² ), and obtaining each service Connect the shareable relationship of backup resources, and calculate the maximum recovery time RT _i ¹ and RT _i ² corresponding to the first backup path BP _i ¹ and the second backup path BP _i ² in each business connection;

103. For any service connection c _i , if BP _i ² can be shared, and RT _i ² ≤ RT ₀ (RT ₀ is the service recovery time threshold), then set BP _i ² as shared configuration; otherwise, set BP _i ² as Dedicated configuration; if BP _i ¹ can be shared, and RT _i ¹ ≤ RT ₀ , then set BP _i ¹ as a shared configuration, otherwise, set BP _i ¹ as a dedicated configuration;

104. On any link l in the optical network G(N,L), allocate service bandwidth for all working paths WP _i passing through the link is the original service bandwidth requirement, l∈L, and adopts the resource optimization ILP model to calculate and obtain the optimal allocated bandwidth on the corresponding first backup path BP _i ¹ and the second backup path BP _i ² Among them, k is the service level, β _k is the backup channel bandwidth allocation ratio, and 0≤β ₁ ≤β ₂ ≤β ₃ ...≤β _k ≤1;

105. On any link l in the optical network G(N,L), allocate shared backup bandwidth resources for related service connections, satisfying and Among them, l∈L, k and k' represent different service levels, β _k' and β _k' represent the backup path bandwidth allocation ratios of different levels of services, and are the backup bandwidths of connections _ci and _ci' passing through link l respectively.

106. On any link l in the optical network G(N,L), allocate dedicated backup bandwidth resources for relevant business connections, satisfying Among them, l∈L, k and k' represent different service levels, β _k' and β _k' represent the backup path bandwidth allocation ratios of different levels of services, and are the backup bandwidths of connections _ci and _ci' passing through link l respectively.

The above embodiments should be understood as only for illustrating the present invention but not for limiting the protection scope of the present invention. After reading the contents of the present invention, skilled persons can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims (3)

1. in a kind of WDM optical network, based on the hybrid protection method sensitive to recovery time, it is characterized in that, comprising the following steps: 101. Use the Dijkstra algorithm in the optical network G(N,L) to find three separate paths for each service connection c _i , namely: the working path WP _i , the first backup path BP _i ¹ and the second backup path BP _i ² , and BP _i ² is not shorter than BP _i ¹ , 1≤i≤|C|, C represents the set of all business connections in the network, N represents the set of all nodes in the network, L represents all bidirectional links in the network gather; 102. Comparing the mutual positional relationship between any paths of all service connections in the optical network G(N,L), wherein any paths include the working path WP _i , the first backup path BP _i ¹ and the second backup path BP _i ² , and obtain the The sharable relationship of the backup resources of the business connection, and calculate the maximum recovery time RT _i ¹ and RT _i ² corresponding to the first backup path BP _i ¹ and the second backup path BP _i ² in each business connection; 103. According to the sharable relationship of the first and second backup paths, they are shared or dedicatedly configured in combination with the service recovery time threshold; 104. On any link l in the optical network G(N,L), allocate service bandwidth for all working paths WP _i passing through the link is the original service bandwidth requirement, l∈L, and adopts the resource optimization ILP model to calculate and obtain the optimal allocated bandwidth on the corresponding first backup path BP _i ¹ and the second backup path BP _i ² Among them, k is the service level, β _k is the backup channel bandwidth allocation ratio, and 0≤β ₁ ≤β ₂ ≤β ₃ ...≤β _k ≤1; 105. On any link l in the optical network G(N,L), allocate shared backup bandwidth resources for related service connections, satisfying and Among them, l∈L, k and k' represent different service levels, β _k' and β _k' represent the backup path bandwidth allocation ratios of different levels of services, and are respectively the backup bandwidths of connections _ci and _ci′ passing through link l; 106. On any link l in the optical network G(N,L), allocate dedicated backup bandwidth resources for relevant business connections, satisfying Among them, l∈L, k and k' represent different service levels, β _k' and β _k' represent the backup path bandwidth allocation ratios of different levels of services, and are the backup bandwidths of connections _ci and _ci' passing through link l respectively. 2. in the WDM optical network according to claim 1, based on the recovery time-sensitive hybrid protection method, it is characterized in that step 102 judges whether the backup resource can be shared according to the positional relationship of the path: 1) Rule one Positional relationship: or: or or Sharing relationship: 2) Rule two Positional relationship: Sharing relationship: 3) Rule three Positional relationship: or: or or Sharing relationship: 4) Rule Four Positional relationship: Sharing relationship: 5) Rule five Positional relationship: any other positional relationship other than the above positional relationship; Sharing relationship: 3. In the WDM optical network according to claim 1 or 2, the hybrid protection method based on recovery time sensitivity is characterized in that, step 103 is based on the shareable relationship of the first and second backup paths, combined with the service recovery time threshold Make shared or private configuration settings, including: For any service connection c _i , if the second backup path BP _i ² can be shared, and RT _i ² ≤ RT ₀ , and RT ₀ is the service recovery time threshold, then set BP _i ² as shared configuration, otherwise, set BP _i ² is a dedicated configuration; if BP _i ¹ can be shared and RT _i ¹ ≤ RT ₀ , then set BP _i ¹ as a shared configuration; otherwise, set BP _i ¹ as a dedicated configuration.