WO2012024965A1 - Method and system for triggering re-optimization - Google Patents

Abstract

A method and system for triggering re-optimization are disclosed in the invention. In the method, a border node in an area sends a message, which carries information indicating that traffic engineering (TE) information changes, to the head node of a label switched path (LSP) which passes through the area (S102); after receiving the message, the head node determines to perform the re-optimization on the LSP (S104). The invention can enable the head node to obtain the change of the TE information and to perform the re-optimization on the LSP, thereby better optimizing the network traffic.

Description

TECHNICAL FIELD The present invention relates to the field of communications, and in particular to a method and system for triggering re-optimization. BACKGROUND OF THE INVENTION Internet Protocol (IP)/Multi Protocol Label Switch (MPLS) for data communication, Multi Protocol Label Switch-Transport Profile (Transport Profile) Traffic MPLS (TP), General Multi Protocol Label Switch (GMPLS), etc. Traffic Management Protocol (TE) is used for traffic engineering protocol (TEVP-TE), and flooding of TE information (including bandwidth/delay/protection) is based on traffic engineering. (Open Shortest Path First-TE, OSPF-TE for short). The flooding of TE information is generally based on the Link State Advertisement (Opaque LSA). The scope of the advertisement is limited to the area. ) Internal. Therefore, when the TE information of the network changes (for example, the bandwidth resource is released on the link), only the nodes in the area can detect the change of the TE information. The nodes outside the area cannot be perceived by the routing protocol. When an LSP crosses an area, the ingress node of the LSP can locally optimize the traffic through the change of the Traffic Engineering Data (TED). However, local optimization has two disadvantages: (1) local optimization is not globally optimal; (2) local optimization is unsuccessful. For this reason, re-optimization (also known as global traffic optimization) is a good choice. The re-optimization technique is to establish a better path while keeping the current path available, and then delete the current path and use the newly established path to transmit data. Currently, in the field of optical transmission, the solution of re-optimization can be implemented by using Path Computation Element (PCE, for details, please refer to RFC4655), but PCE is not deployed in the traditional data network. Because the nodes in one area of the data network cannot learn the dynamic changes of the information in other areas, the LSP re-optimization cannot be triggered. SUMMARY OF THE INVENTION The present invention provides a method and system for triggering re-optimization, because the node in one area of the data network cannot learn the dynamic change of TE information in other areas, so that the re-optimization of the LSP cannot be triggered. Solve at least one of the above problems. According to an aspect of the present invention, a re-optimized triggering method is provided. The method for triggering the re-optimization according to the present invention includes: the border node of the area sends a message to the head node of the LSP that traverses the area, where the message carries information for indicating that the TE information changes; after the first node receives the message, Determine to perform re-optimization on the LSP. The above information includes: information indicating a current path error, or information indicating that the current path is available but there is a path superior to the path. The above information is indicated by the error value Error Value. The above message includes: Notifying Notify message or path error PathErr message. When the message is a Notify message, before the border node sends the message to the head node, the method further includes: the border node receiving the path Path message carrying the notification request Notify Request object from the head node; then the border node sends the message to the head node The message includes: The border node sends the Notify message directly to the head node through IP routing. When the foregoing message is a PathErr message, the sending, by the border node, the message to the head node includes: the border node transmitting the PathErr message to the head node megabytes on the LSP. The above area includes one of the following: a routing autonomous domain, a partition in the routing autonomous domain. According to another aspect of the invention, a re-optimized trigger system is provided. The re-optimized triggering system according to the present invention includes: a border node located at a boundary of the area, configured to send a message to a head node of the LSP passing through the area, wherein the message carries information for indicating that the TE information is changed; Set to perform re-optimization of the LSP after receiving the message. The foregoing border node is further configured to: after receiving the path Path message carrying the notification request Notify Request object from the first node, when the message is the Notify message, the Notify message is directly sent to the first node by using the IP route. The boundary node is further configured to transmit the PathErr message hop by hop to the first node on the LSP when the message is a path error PathErr message. According to the present invention, when the border node of the area learns that the TE information changes, the message is sent to the head node of the label switched path LSP that passes through the area, where the message carries information indicating that the TE information is changed. The first node learns the change of the TE information, and determines whether to perform re-optimization on the LSP. In the related art, the node in one area of the data network cannot learn the dynamic change of the TE information in other areas, so that the problem of re-optimization of the network traffic cannot be triggered, and the first node can learn the change of the TE information, and the LSP is heavy. Optimized to better optimize network traffic. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are set to illustrate,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In the drawings: FIG. 1 is a flowchart of a re-optimization triggering method in accordance with an embodiment of the present invention; FIG. 2 is a flowchart of a re-optimization triggering method in a preferred embodiment of the present invention; BRIEF DESCRIPTION OF THE DRAWINGS FIG. 4 is a schematic diagram of a second embodiment of the present invention; FIG. 5 is a structural block diagram of a re-optimized trigger system in accordance with an embodiment of the present invention. BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. 1 is a flow chart of a re-optimized triggering method in accordance with an embodiment of the present invention. As shown in FIG. 1 , the method for triggering the re-optimization includes: Step S102: The border node of the area sends a message to the head node of the LSP that passes through the area, where the message carries information for indicating that the TE information changes; S104: After receiving the message, the first node determines to perform re-optimization on the LSP. In the related art, it is solved that the node in one area of the data network cannot learn the dynamic change of the TE information in other areas in the related art, so that the re-optimization of the LSP cannot be triggered. Using the above method, the first node can learn the change of the TE information and re-optimize the LSP to better optimize the network traffic. Preferably, the foregoing information includes but is not limited to: information indicating a current path error, or information indicating that the current path is available but there is a path superior to the path. In a preferred implementation process, when the border node of the foregoing area senses that the TE information changes, by transmitting the foregoing information in the message to notify the first node that passes through the area LSP (the node is located in another area), the foregoing information may pass the error value ( Error Value ) to indicate. For example, two new error values can be defined: Error Value=12; Error Value= 13 , where when the Error Value is equal to 12, it indicates the current path error; when Error Value is equal to 13, it indicates that the current path is available but exists Information about paths with better paths. Preferably, the foregoing message includes but is not limited to: notifying a Notify message or a path error PathErr message. In a preferred implementation process, the path state request transmitted between the nodes is implemented by a Path message in a Resource Reservation Protocol (RSVP) message, and the reservation state request passed between the nodes passes the Resv in the RSVP message. The message is implemented. When the Resv message is processed, the label is allocated, the resource is reserved, and the LSP is established. When an LSP fails or an error is encountered during the establishment, the PathErr message in the RSVP message can be used to hop-by-hop the first node of the LSP, or the Notify message can be used to notify the first node of the LSP. After receiving the foregoing message, the first node determines whether to perform re-optimization on the LSP according to the actual situation of the current LSP. The method further includes: before the message is a Notify message, before the border node sends the message to the first node, the method further includes: the border node receiving the path Path message carrying the notification request Notify Request object from the first node; The node sending the message to the head node may include the following processing: The border node sends the Notify message directly to the head node through the IP route. When the message is a PathErr message, the sending of the message by the border node to the head node in the foregoing step S102 includes the following processing: The border node transmits the PathErr message hop by hop on the LSP to the first node. That is, the premise that the border node sends the Notify message to the head node is that the border node has previously received the Path message carrying the Notify Request object sent by the head node; if not, the border node can only A PathErr message is sent to the head node, and the message can be delivered to the head node only one hop, and the speed is >1"mans compared with the Notify message. Preferably, the message further carries the following information: The path of the first node re-established The information about the nodes to be included is required. The above preferred embodiment can be specifically referred to FIG. 2. Fig. 2 is a flowchart of a re-optimized triggering method according to a preferred embodiment of the present invention. As shown in Fig. 2, the trigger of the re-optimization The method includes the following steps: Step S202: The TE information of an area in the data network changes; Step S204: The boundary node in the area senses that the TE information changes; Step S206: The boundary node of the area indicates that the TE information is generated. The changed information is sent to the first node of the LSP passing through the area; Step S208: The first node determines to perform re-optimization on the LSP, and selects a traffic optimization scheme based on a certain strategy. Preferably, the foregoing area may include one of the following: Domain, routing, partition in the autonomous domain. The following describes the cross-routing autonomous domain in conjunction with the examples in Figure 3 and Figure 4, respectively. The global traffic optimization (re-optimization) trigger mode of the area and the cross-route autonomous domain (AS). Figure 3 is a schematic diagram of the first embodiment of the present invention. As shown in Figure 3, the routing autonomous domain 1 (AS 1) There are three zones i or, are areal, area0, area2. The TE metrics values of the links in area2 are as follows: R6-R8 is 9, R8-R11 is 6, R7-R9 is 11, R8-R9 is 13, R9-R10 is 10, and R10-R11 is 8. The metrics of these TEs are generally assigned by the administrator, which comprehensively reflect information about link delay, jitter, etc. There is an LSP1 between R1 and R11 (R1- R3-R5-R7-R9-R10-R11), the bandwidth is 1G; the Unreserved Bandwidth value between R8-R11 is 600M. Assume that the bandwidth of an LSP passing through R8-R11 is at time t0. After being released, the Unreserved Bandwith value between R8 and R11 is increased to 1.2 G. The change of this TE information is flooded by OSPF-TE, and finally the boundary nodes R6 and R7 of the area are reached.

R6 (The R7 node is judged the same as the R6 node and will not be described again.) After the node learns that the Unreserved Bandwith of the link between R8 and R11 is 1.2G, it determines whether there is a cross-domain LSP passing through the area. If not, ignore it, if any, check the bandwidth of these LSPs; if the bandwidth of these LSPs is greater than 1.2G, Ignore; if the bandwidth of these LSPs is less than or equal to 1.2G, then the node sends a Notify or PathErr message with the newly defined Error Value value to the head nodes of these LSPs. Of course, if the number of LSPs with bandwidth less than 1.2G is still large, the node can choose to transmit only on a part of the LSP based on the local policy. In this example, the bandwidth 1G of LSP1 is less than 1.2G, so the R7 node wants to send a Notify message or a PathErr message to the first node R1 of LSP1. It should be noted that the premise that the R7 node can send the Notify message to the R1 node is that the R7 node has previously received the Path message carrying the Notify Request object sent by R1; if not, the R7 node can only send the PathErr message to the R1 node. The message can only be passed to the R1 node one hop and one hop, which is '1' mile faster than the Notify message.

After receiving the Notify or PathErr message, the R1 node determines whether the LSP needs to be re-optimized. For example, if the LSP is very sensitive to delay or jitter, the LSP needs to be re-optimized according to the SLA protocol signed with the user. If the LSP is not sensitive to delay or jitter, then no re-optimization is needed. Figure 4 is a schematic diagram of Example 2 of the present invention. As shown in Figure 4, two routing autonomous domains (AS) are shown, which are AS 1 and AS 2, respectively. The TE metrics values of the links in AS2 are as follows: R6-R8 is 9, R8-R11 is 6, R7-R9 is 11, R8-R9 is 13, R9-R10 is 10, R10-R11 is 8; The metrics are generally assigned by the administrator and comprehensively reflect information about the link delay, jitter, and so on. There is an LSP1 ( R1-R3-R5-R7-R9-R10-R11) between R1 and R11 with a bandwidth of 1G. The Unreserved Bandwidth value between R8 and R11 is 600M. Assume that at time t0, the bandwidth of an LSP passing through R8-R11 is released, then the Unreserved Bandwith value between R8-R11 is increased to 1.2G. This TE information change is flooded by OSPF-TE, and eventually reaches the boundary nodes R6 and R7 of the area.

R6 (The R7 node is judged the same as the R6 node and will not be described again.) After the node learns that the Unreserved Bandwith of the link between R8 and R11 is 1.2G, it determines whether there is a cross-domain LSP passing through the area. If not, ignore it; if yes, check the bandwidth of these LSPs; if the bandwidth of these LSPs is greater than 1.2G, ignore it; if the bandwidth of these LSPs is less than or equal to 1.2G, then the node sends the insertion to the first node of these LSPs. A Notify or PathErr message for the newly defined Error Value value. Of course, if the number of LSPs with bandwidth less than 1.2G is still large, the node can choose to transmit only on a part of the LSP based on the local policy. In this example, the bandwidth 1G of LSP1 is less than 1.2G, so the R7 node wants to send a Notify message or a PathErr message to the first node R1 of LSP1. It should be noted that the premise that the R7 node can send the Notify message to the R1 node is that the R7 node has previously received the Path message carrying the Notify Request object sent by R1; if not, the R7 node can only send the PathErr message to the R1 node. The message can only be passed to the R1 node one hop and one hop, which is '1' mile faster than the Notify message.

After receiving the Notify or PathErr message, the R1 node determines whether the LSP needs to be re-optimized. For example, if the LSP is very sensitive to delay or jitter, the LSP needs to be re-optimized according to the SLA protocol signed by the user; if the LSP is not sensitive to delay or jitter, then no re-optimization is needed. It can be seen from FIG. 3 and FIG. 4 that the re-optimized triggering scheme provided by the embodiment of the present invention can be applied not only to the scenario of the cross-route autonomous domain, but also to the scenario of the cross-routing autonomous domain, and thus the application scope is wide. FIG. 5 is a structural block diagram of a re-optimized trigger system in accordance with an embodiment of the present invention. As shown in FIG. 5, the re-optimized trigger system includes: a boundary node 50 of the area and a head node 52 of the LSP passing through the area. The border node 50 is located at a boundary of the area, and is configured to send a message to a head node of the label switching path LSP that passes through the area, where the message carries information for indicating that the TE information changes; the head node 52 is set to receive After the message, it is determined to perform re-optimization on the LSP. Using the above method, the first node can learn the change of the TE information and re-optimize the LSP to better optimize the network traffic. The foregoing area may include, but is not limited to, one of the following: a routing autonomous domain, a partition in a routing autonomous domain. For example, as shown in FIG. 4, when the foregoing area is a routing autonomous domain, for the routing autonomous domain AS2, the boundary node may be R6 or R7. The LSP that traverses this area (AS2) is two, and the first node is R1. The foregoing information includes but is not limited to: information indicating a current path error, or information indicating that the current path is available but there is a path superior to the path. Wherein, the above information can be represented by an error value (Error Value). For example, you can define two new error values: Error Value=12; Error Value=13 , where, when Error Value is equal At 12 o'clock, the current path error is indicated; when Error Value is equal to 13, information indicating that the current path is available but there is a better path than the path. Preferably, the foregoing message may be a Notify message or a path error PathErr message or the like. Preferably, if the message is a Notify message, the border node 50 is further configured to: after the message is a Notify message, after receiving the path Path message carrying the notification request Notify Request object from the head node, notify Notify through the IP route. The message is sent directly to the head node. Preferably, if the message is a PathErr message, the border node 50 is further configured to transmit the PathErr message to the head node on a LSP by megabyte. That is, the premise that the border node sends the Notify message to the head node is that the border node has previously received the Path message carrying the Notify Request object sent by the head node; if not, the border node can only send the PathErr message to the head node, the message can only be The hop is delivered to the head node, and the speed is >1"manier than the Notify message. In a preferred implementation, the message may further carry information about the node to be included in the path re-established by the head node 52. For a preferred embodiment of the combination of the boundary node 50 and the head node 52, reference may be made to the description of FIG. 2 to FIG. 4, and details are not described herein. In summary, the above embodiments provided by the present invention, through a simple signaling protocol The extension may be performed to enable the first node of the inter-area LSP to learn the change of the TE information, thereby determining to perform re-optimization on the LSP. The related art is that the node in one area of the data network cannot learn the dynamic change of the TE information in other areas. Therefore, the problem of re-optimization of network traffic cannot be triggered, and thus the first node can be optimized for LSPs. Solution, the network traffic is better optimized. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, which can be concentrated on a single computing device, or Distributed over a network of computing devices, optionally, they may be implemented in program code executable by a computing device, such that they may be stored in a storage device for execution by the computing device, and in some cases The steps shown or described may be performed in a different order than that herein, or they may be separately fabricated into individual integrated circuit modules, or a plurality of the modules or steps may be implemented as a single integrated circuit module. The invention is not limited to any particular combination of hardware and software. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Claims

Claim 1. A re-optimized triggering method, including:  The border node of the area sends a message to the head node of the label switching path LSP that passes through the area, where the message carries information indicating that the TE information of the traffic engineering changes.  After receiving the message, the head node determines to perform re-optimization on the LSP. 2. The method according to claim 1, wherein the information comprises: information indicating a current path error, or information indicating that a current path is available but a path superior to the path exists. 3. The method of claim 1, wherein the information is represented by an error value of Error Value. 4. The method according to claim 1, wherein the message comprises: a Notify message or a Path Error PathErr message. 5. The method according to claim 4, wherein when the message is a Notify message,  Before the sending, by the border node, the message to the first node, the method further includes: the border node receiving, by the first node, a path Path message carrying a notification request Notify Request object;  And sending, by the border node, the message to the head node includes: sending, by the border node, the Notify message to the head node by using an IP route. The method according to claim 4, wherein, when the message is a PathErr message, the sending, by the border node, the message to the head node comprises: the border node transmitting a hop by hop on the LSP The PathErr message is described up to the first node. The method according to any one of claims 1 to 6, wherein the area comprises one of: a routing autonomous domain, a partition in a routing autonomous domain. 8. - A variety of optimized trigger systems, including:  a boundary node, located at a boundary of the area, configured to send a message to a head node of the label switched path LSP that passes through the area, where the message carries information indicating that the traffic engineering TE information changes;  The head node is configured to, after receiving the message, determine to perform re-optimization on the LSP. 9. The system according to claim 8, wherein the border node is further configured to receive a path Path message carrying the notification request Notify Request object from the head node when the message is a Notify message. Thereafter, the Notify message is sent directly to the head node by IP routing. 10. The system according to claim 9, wherein the border node is further configured to transmit the PathErr message hop by hop on the LSP until the head node when the message is a path error PathErr message.