CN116437411A - Automatic path calculation and path selection method for realizing SRv-TE-Policy based on hybrid Policy - Google Patents

Abstract

The present invention relates to the field of network transport technology, and discloses an automatic route calculation and route selection method based on a mixed strategy to realize SRv6-TE-Policy, including a total backbone network, soft slice and automatic drainage design, intelligent traffic scheduling implementation mechanism, SRv6 ‑TE‑Policy application, building SRv6‑TE‑Policy model, SRv6‑TE‑Policy path calculation, SRv6‑TE‑Policy drainage; the total backbone network is divided into core backbone network and first-level backbone network, and the total backbone network includes control Protocol definition among routers, repeaters, controllers and repeaters. The automatic path calculation and selection method of SRv6‑TE‑Policy is realized based on the hybrid policy, and the total backbone network is designated for network route calculation and route selection. Soft slicing and automatic traffic diversion divide network traffic and adjust networks with different paths Transmission pressure, rationally divide different traffic information through intelligent traffic scheduling, build SRv6-TE-Policy model, switch transmission paths for network information transmission failures to avoid affecting network use, and SRv6-TE-Policy guides network traffic .

Description

An automatic route calculation and route selection based on hybrid policy for SRv6-TE-Policy method

technical field

The invention relates to the technical field of network transmission, in particular to an automatic path calculation and path selection method for realizing SRv6-TE-Policy based on a mixed strategy.

Background technique

The 5G network supports flexible business scenarios. Customers have different requirements for the network. For example, the to B application in the uRLLC scenario hopes to provide a network path with low latency and low jitter, while the to C application in the eMBB scenario requires a low packet loss rate. High bandwidth channel. The original IPRAN bearer network configures the routing protocol on the device side (such as the OSPF protocol IS-IS protocol, etc.), and the device calculates the network path that the service traffic needs to pass through according to the shortest path SPF algorithm of the protocol. This method is used in 4G It is widely used in 5G NSA and 5GN SA networks, but it cannot meet the needs of flexible traffic scheduling according to user application requirements in the 5G SA network to B business scenario. Therefore, this application proposes a hybrid strategy based on SRv6-TE-Policy Automatic route calculation and route selection methods to deal with problems such as failures in the network transmission process.

Contents of the invention

(1) Solved technical problems

Aiming at the deficiencies of the prior art, the present invention provides an automatic route calculation and route selection method based on a hybrid policy to realize SRv6-TE-Policy, which has the advantages of traffic scheduling and troubleshooting during network transmission, and solves many current network problems. Slow delivery speeds and failures affect network usage.

(2) Technical solution

An automatic path calculation and path selection method based on a hybrid strategy to realize SRv6-TE-Policy, including the overall backbone network, soft slice and automatic traffic diversion design, intelligent traffic scheduling implementation mechanism, SRv6-TE-Policy application, and construction of SRv6-TE-Policy Policy model, SRv6-TE-Policy path calculation, SRv6-TE-Policy drainage;

The total backbone network is divided into a core backbone network and a first-level backbone network. The total backbone network includes controllers, transponders, and protocol delineation between controllers and transponders;

The automatic path calculation and path selection method of SRv6-TE-Policy includes the following steps:

S101. Designate the general backbone network to perform network route calculation and route selection;

S102, soft slicing and automatic traffic diversion divide network traffic;

S103. Reasonably divide different traffic information through intelligent traffic scheduling;

S104. Build an SRv6-TE-Policy model;

S105. Use SRv6-TE-Policy to switch the transmission path when a failure occurs during network information transmission;

S106, SRv6-TE-Policy diverts and directs network traffic.

Preferably, the controller is equivalent to a central processing unit, grasps information such as the topology of the entire network, real-time traffic, SRv6SID, etc., and is responsible for translating user intentions into SRv6 Policy and sending it to the transponder through the southbound protocol;

The forwarder is responsible for route calculation and SRv6 Policy encapsulation and forwarding. On the one hand, it needs to run ISIS to calculate the Underlay route and SID, run BGP EVPN to calculate the Overlay route and SID, and iterate the Overlay route to the appropriate SRv6 Policy. Report the Underlay network and SRv6 Policy status;

The agreement between the controller and the forwarder is to run multiple southbound protocols between the controller and the forwarder, the most important being BGP-LS and BGP SR Policy, the former is responsible for converting the link state of the forwarder into BGP -LS messages and the status of the SRv6 Policy are reported to the controller, and the latter sends the SRv6 Policy programmed by the controller to the forwarder.

Preferably, soft slicing and automatic traffic diversion are designed. The backbone network adopts soft slicing technology based on SRv6 Policy. Compared with conventional QoS policies, network slicing can better isolate and protect different types of business traffic. Taking the core backbone network as an example, three types of soft slices (STORAGE, INTRANET, and EXTRANET) are defined, that is, Color1 SRv6 Policy, which is used to carry three types of service VPN and traffic of storage, intranet, and extranet. It is possible to avoid traffic conflicts. Each slice is divided into two Color2 SRv6 policies with high and low priorities, which are used to carry business traffic of different service levels within the same VPN.

The PE node of the backbone network is also the core of the access network, providing external access to the service domain and user domain of the head office and branch. The ingress PE adopts the two-level fine flow scheduling technology of DSCP+SRv6 Policy. First, according to the <color, nexthop> matches the soft slice <color1, endpoint>, and then maps to the Color2 SRv6 Policy in the soft slice according to the DSCP value of the user packet for intelligent forwarding, providing users with more detailed traffic scheduling and service experience.

Preferably, the implementation mechanism of intelligent traffic scheduling. The SDN WAN controller is the brain of intelligent traffic scheduling decisions. It will calculate the specific forwarding path of each slice policy according to the user's intention, that is, the segment list, and send it to the head-end PE node, and expand it through SRH The head carries route information, just like a car with its own navigation system, the entire route is known as soon as it is on the road, and the intermediate nodes do not need to maintain any state information, which can support any size of network scale. The current backbone network uses UCMP (Uequal Cost Multiple Path) intelligent scheduling mechanism.

Preferably, the intelligent traffic scheduling implementation mechanism uses the SD WAN controller to monitor the network traffic, and carries the network transmission path information through the SRH extension header.

Preferably, the model for constructing SRv6-TE-Policy determines the three elements of independence, reliability, and stability. Independence: any business can call this interface without paying attention to the internal details of the system. SRv6Policy encapsulates all the details internally. When a business uses SRv6 Policy, it only needs to find this interface;

Reliability: In the case where the interface has been published, the interface should be responsible to the contract. No matter how many services call this interface, SRv6 Policy needs to guarantee the services promised in the contract, which requires SRv6 Policy to have the ability to flexibly scale network resources;

Stability: The interface needs to be invariable. During the SRv6 Policy life cycle, regardless of the change of the network topology, the change of the service itself or the change of the path, the Binding SID needs to be kept unchanged as much as possible.

Preferably, the path calculation of SRv6-TE-Policy includes three parts: the static specified path, the path calculation of the head node, and the path calculation of the controller. When the specified link fails during the transmission of network information, the switching of the transmission path is performed. The head node first The head node uses the TE information carried by the IGP and the IGP link state information to form a TEDB, and then based on the CSPF algorithm, it calculates the satisfaction conditions according to constraints such as bandwidth, delay, SRLG (Shared Risk Link Group, shared risk link group) and disjoint paths. path, and install the corresponding SRv6 Policy to guide the forwarding; the path calculation of the controller can obtain the global topology and TE information through BGP-LS, so the calculation of SRv6 Policy based on the controller can realize the optimization of global traffic, and statically specify the path The path calculation method with the head node can only realize the optimal path calculation in the IGP domain. In addition, the path calculation of the controller can also support bandwidth reservation and priority preemption, which can better support TE.

Preferably, the SRv6-TE-Policy diversion deploys the SRv6 Policy after the head node, and also needs to complete the diversion work to guide the traffic to the SRv6Policy.

Compared with the prior art, the present invention provides an automatic path calculation and path selection method based on a hybrid strategy to realize SRv6-TE-Policy, which has the following beneficial effects:

1. The automatic route calculation and route selection method of SRv6-TE-Policy is implemented based on the hybrid strategy, and the total backbone network is designated for network route calculation and route selection. Soft slicing and automatic traffic diversion divide network traffic and adjust different paths According to the network transmission pressure, different traffic information is reasonably divided through intelligent traffic scheduling, and the SRv6-TE-Policy model is constructed. The SRv6-TE-Policy is used to switch the transmission path to avoid affecting the use of the network when faults occur during network information transmission. SRv6 -TE-Policy diverts and guides network traffic.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

An automatic path calculation and path selection method based on a hybrid strategy to realize SRv6-TE-Policy, including the overall backbone network, soft slice and automatic traffic diversion design, intelligent traffic scheduling implementation mechanism, SRv6-TE-Policy application, and construction of SRv6-TE-Policy Policy model, SRv6-TE-Policy path calculation, SRv6-TE-Policy drainage;

The total backbone network is divided into a core backbone network and a first-level backbone network. The total backbone network includes controllers, transponders, and protocol delineation between controllers and transponders;

The automatic path calculation and path selection method of SRv6-TE-Policy includes the following steps:

S101. Designate the general backbone network to perform network route calculation and route selection;

S102, soft slicing and automatic traffic diversion divide network traffic;

S103. Reasonably divide different traffic information through intelligent traffic scheduling;

S104. Build an SRv6-TE-Policy model;

S105. Use SRv6-TE-Policy to switch the transmission path when a failure occurs during network information transmission;

S106, SRv6-TE-Policy diverts and directs network traffic.

Further, the controller is equivalent to the central processing unit, which grasps information such as the topology of the entire network, real-time traffic, and SRv6SID, and is responsible for translating user intentions into SRv6 Policy and sending it to the transponder through the southbound protocol;

The forwarder is responsible for route calculation and SRv6 Policy encapsulation and forwarding. On the one hand, it needs to run ISIS to calculate the Underlay route and SID, run BGP EVPN to calculate the Overlay route and SID, and iterate the Overlay route to the appropriate SRv6 Policy. Report the Underlay network and SRv6 Policy status;

The agreement between the controller and the forwarder is to run multiple southbound protocols between the controller and the forwarder, the most important being BGP-LS and BGP SR Policy, the former is responsible for converting the link state of the forwarder into BGP -LS messages and the status of the SRv6 Policy are reported to the controller, and the latter sends the SRv6 Policy programmed by the controller to the forwarder.

Further, soft slicing and automatic traffic diversion design. The backbone network adopts soft slicing technology based on SRv6 Policy. Compared with conventional QoS policies, network slicing can better isolate and protect different types of business traffic. Take the core backbone network as an example , defines three types of soft slices, namely STORAGE, INTRANET, and EXTRANET, namely Color1 SRv6Policy, which are used to carry three types of service VPNs and traffic of storage, intranet, and extranet. The paths between slices are independent, and traffic conflicts are avoided as much as possible. Each The slice is divided into two Color2 SRv6 policies with high and low priorities, which are used to carry business traffic of different service levels within the same VPN.

The PE node of the backbone network is also the core of the access network, providing external access to the service domain and user domain of the head office and branch. The ingress PE adopts the two-level fine flow scheduling technology of DSCP+SRv6 Policy. First, according to the <color, nexthop> matches the soft slice <color1, endpoint>, and then maps to the Color2 SRv6 Policy in the soft slice according to the DSCP value of the user packet for intelligent forwarding, providing users with more detailed traffic scheduling and service experience.

Further, the implementation mechanism of intelligent traffic scheduling. The SDN WAN controller is the brain of intelligent traffic scheduling decisions. It will calculate the specific forwarding path of each slice policy according to the user's intention, that is, the segment list, and send it to the head-end PE node, and expand it through SRH The head carries route information, just like a car with its own navigation system, the entire route is known as soon as it is on the road, and the intermediate nodes do not need to maintain any state information, which can support any size of network scale. The current backbone network uses UCMP (Uequal Cost Multiple Path) intelligent scheduling mechanism.

Furthermore, the intelligent traffic scheduling implementation mechanism uses the SD WAN controller to monitor the network traffic, and carries the network transmission path information through the SRH extension header.

Furthermore, the model for constructing SRv6-TE-Policy determines the three elements of independence, reliability, and stability. Independence: Any business can call this interface without paying attention to the internal details of the system. SRv6Policy encapsulates all the details internally. When a business uses SRv6 Policy, it only needs to find this interface;

Reliability: In the case where the interface has been published, the interface should be responsible to the contract. No matter how many services call this interface, SRv6 Policy needs to guarantee the services promised in the contract, which requires SRv6 Policy to have the ability to flexibly scale network resources;

Stability: The interface needs to be invariable. During the SRv6 Policy life cycle, regardless of the change of the network topology, the change of the service itself or the change of the path, the Binding SID needs to be kept unchanged as much as possible.

Furthermore, SRv6-TE-Policy path calculation includes three parts: static specified path, head node path calculation, and controller path calculation. When the specified link fails during network information transmission, the transmission path will be switched. The head node first The head node uses the TE information carried by the IGP and the IGP link state information to form a TEDB, and then based on the CSPF algorithm, it calculates the satisfaction conditions according to constraints such as bandwidth, delay, SRLG (Shared Risk Link Group, shared risk link group) and disjoint paths. path, and install the corresponding SRv6 Policy to guide the forwarding; the path calculation of the controller can obtain the global topology and TE information through BGP-LS, so the calculation of SRv6 Policy based on the controller can realize the optimization of global traffic, and statically specify the path The path calculation method with the head node can only realize the optimal path calculation in the IGP domain. In addition, the path calculation of the controller can also support bandwidth reservation and priority preemption, which can better support TE.

Furthermore, SRv6-TE-Policy diversion deploys SRv6 Policy behind the head node, and the diversion work needs to be completed to guide traffic to SRv6Policy.

Working principle: When the automatic route calculation and route selection method of SRv6-TE-Policy is implemented based on the hybrid policy, the total backbone network is designated for network route calculation and route selection, and soft slicing and automatic traffic diversion are used to divide network traffic. , adjust the network transmission pressure of different paths, rationally divide different traffic information through intelligent traffic scheduling, build the SRv6-TE-Policy model, and use SRv6-TE-Policy to switch the transmission path to avoid affecting the network when faults occur during network information transmission SRv6-TE-Policy diverts and directs network traffic.

Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that various changes, modifications and substitutions can be made to these embodiments without departing from the principle and spirit of the present invention. and modifications, the scope of the invention is defined by the appended claims and their equivalents.

Claims (8)

1. An automatic path calculation and path selection method based on a hybrid strategy to realize SRv6-TE-Policy, which is characterized in that it includes the overall backbone network, soft slice and automatic traffic diversion design, intelligent traffic scheduling implementation mechanism, and SRv6-TE-Policy application , Construct SRv6-TE-Policy model, SRv6-TE-Policy path calculation, SRv6-TE-Policy drainage; The overall backbone network is divided into a core backbone network and a first-level backbone network, and the overall backbone network includes a controller, a transponder, and a protocol definition between the controller and the transponder; The automatic path calculation and path selection method of the SRv6-TE-Policy includes the following steps: S101. Designate the general backbone network to perform network route calculation and route selection; S102, soft slicing and automatic traffic diversion divide network traffic; S103. Reasonably divide different traffic information through intelligent traffic scheduling; S104. Build an SRv6-TE-Policy model; S105. Use SRv6-TE-Policy to switch the transmission path when a failure occurs during network information transmission; S106, SRv6-TE-Policy diverts and directs network traffic. 2. A method for automatic route calculation and route selection based on a hybrid strategy to realize SRv6-TE-Policy according to claim 1, characterized in that: the controller is equivalent to a central processing unit, and grasps the topology of the entire network and real-time traffic , SRv6 SID information; the forwarder is responsible for route calculation and SRv6 Policy encapsulation and forwarding; the agreement between the controller and the forwarder is to run a southbound protocol between the controller and the forwarder. 3. A method for automatic route calculation and route selection based on a hybrid policy to realize SRv6-TE-Policy according to claim 1, characterized in that: the soft slice and automatic traffic drainage design backbone network adopts a soft SRv6 Policy-based Slicing technology, the PE node of the backbone network is also the core of the access network, providing external access to the service domain and user domain of the head office and branches. 4. An automatic route calculation and route selection method based on a hybrid policy to realize SRv6-TE-Policy according to claim 1, characterized in that: the SDN WAN of the intelligent traffic scheduling implementation mechanism will calculate each slice according to user intentions The specific forwarding path of the policy is the segment list, which is delivered to the headend PE node, and the path information is carried through the SRH extension header. 5. A method for automatic route calculation and route selection based on a hybrid strategy to realize SRv6-TE-Policy according to claim 1, characterized in that: the intelligent traffic scheduling implementation mechanism uses SD WAN controllers to monitor network traffic , and carry the network transmission path information through the SRH extension header. 6. A method for automatic route calculation and route selection based on a hybrid strategy to realize SRv6-TE-Policy according to claim 1, characterized in that: the model for constructing SRv6-TE-Policy determines independence, reliability, Three elements of stability. 7. A method for automatic route calculation and route selection based on a hybrid policy to realize SRv6-TE-Policy according to claim 1, characterized in that: said SRv6-TE-Policy route calculation includes static specified paths, head node calculations The path and the controller calculate the path. When the specified link fails during the transmission of network information, the transmission path will be switched. The head node first uses the TE information carried by the IGP and the IGP link status information to form a TEDB , and then based on the CSPF algorithm, according to the constraints of bandwidth, delay, SRLG, and disjoint paths, the path that satisfies the conditions is calculated, and the corresponding SRv6 Policy is installed to guide the forwarding. 8. A method for automatic route calculation and route selection based on a hybrid strategy to realize SRv6-TE-Policy according to claim 1, characterized in that: the SRv6-TE-Policy diversion deploys the SRv6 Policy after the head node, It is also necessary to complete the diversion work and direct the traffic to SRv6Policy.