Not applicable.
Not applicable.
In some networks, such as Multiprotocol Label Switching (MPLS) networks and Generalized MPLS (GMPLS) networks, a Traffic Engineering (TE) Label Switched Path (LSP) can be established by MPLS (or GMPLS) with a path provided by a Path Computation Client (PCC) and a Path Computation Element (PCE). Specifically, the PCC requests a path or route from the PCE, and the PCE computes the path and forwards the computed path information back to the PCC. The path can comprise a plurality of nodes and/or Label Switch Routers (LSRs) and extend from a source node or LSR to a destination node or LSR. In some cases, the path may be a point-to-point (P2P) path, which is computed across single or multiple areas or Autonomous System (AS) domains. In other cases, the path may be a Point-to-Multipoint (P2MP) path from a source node to a plurality of destination nodes, which may be referred to as a tree. However, the mechanisms for requesting and computing the P2MP tree across multiple areas or AS domains using the PCC and PCE are still being developed.
In one embodiment, the disclosure includes a system comprising a plurality of PCEs configured to communicate with an ingress node, jointly compute a core tree for an inter-domain P2MP tree across a plurality of network domains, and independently compute a plurality of sub-trees in at least some of the network domains that have a destination node, wherein the core tree connects the ingress node to a boundary node (BN) in each one of the network domains that have a destination node and each sub-tree connects the BN to a plurality of destination nodes in one of the network domains that have a destination node.
In another embodiment, the disclosure includes a network component comprising at least one processor configured to implement a method comprising obtaining a computation request for an inter-domain P2MP tree from a source node to a plurality of leaf nodes across a plurality of network domains, and computing a core tree that comprises a plurality of paths from the source node to a BN in each of the network domains that comprises at least some of the leaf nodes, calculating a sub-tree for each of the network domains that comprises at least some of the leaf nodes from a BN in the core tree to the leaf nodes, and sending a result of the inter-domain P2MP tree computation to the source node.
In yet another embodiment, the disclosure includes a method comprising computing a core tree from a source node to a BN in each of a plurality of domains that comprise a leaf node, computing a sub-tree from the BN in each of the domains to the leaf node in the same domains, and combining the core tree and the sub-tree in each of the domains to establish inter-domain P2MP tree.
These and other features will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings and claims.
For a more complete understanding of this disclosure, reference is now made to the following brief description, taken in connection with the accompanying drawings and detailed description, wherein like reference numerals represent like parts.
It should be understood at the outset that although an illustrative implementation of one or more embodiments are provided below, the disclosed systems and/or methods may be implemented using any number of techniques, whether currently known or in existence. The disclosure should in no way be limited to the illustrative implementations, drawings, and techniques illustrated below, including the exemplary designs and implementations illustrated and described herein, but may be modified within the scope of the appended claims along with their full scope of equivalents.
Disclosed herein are systems and methods for calculating an inter-domain P2MP path or tree across a plurality of domains. The inter-domain P2MP tree may be computed using a PCC and plurality of PCEs, each of which may be associated with one of the domains. The inter-domain P2MP tree may be established by computing a core tree across the domains and a plurality of sub-trees in each destination domain, which may comprise destination nodes. The core tree may comprise at least one BN in each domain and may be computed based on a Backward Recursive Path Calculation (BRPC) procedure, e.g. using at least some of the PCEs associated with the domains. The sub-tree may be computed in each destination domain using a PCE associated with the domain. The sub-tree may comprise the destination nodes in each destination domain, may be coupled to a BN of the core tree, and may be computed based on a Constrained Shortest Path First (CSPF) procedure. As such, by separately computing the core tree and the sub-trees of the inter-domain P2MP may provide improved LSP computation and may be practical for implementation, e.g. in terms of scalability.
In an embodiment, the label switched network 110 may be a packet switched network, where data traffic may be transported using packets or frames along network paths or routes. The packets may be routed or switched along a TE LSP established by a signaling protocol, such as MPLS or GMPLS, based on a path computed by the PCE and/or developed by the nodes 112. The label switched network 110 may comprise a plurality of nodes 112 coupled to one another using optical, electrical, or wireless links. The label switch network 110 may also comprise a plurality of domains, such as AS domains or interior gateway protocol (IGP) areas, which may each comprise a set of network elements corresponding to the same address management and/or path computational responsibility. The domains may be organized via physical means (e.g. location, connections, etc.) or logical means (e.g. network topology, protocols, communication layers, etc.). The different domains may be coupled to each other and may each comprise some of the nodes 112.
In an embodiment, the nodes 112 may be any devices or components that support transportation of the packets through the label switched network 110. For example, the nodes 112 may include bridges, switches, routers, or various combinations of such devices. The nodes 112 may comprise a plurality of ingress ports for receiving packets from other nodes 112, logic circuitry that determines which nodes 112 to send the frames to, and a plurality of egress ports for transmitting frames to the other nodes 112. In some embodiments, at least some of the nodes 112 may be LSRs, which may be configured to modify or update the labels of the packets transported in the label switched network 110. Further, some of the nodes 112 may be label edge routers (LERs). For example, the nodes 112 at the edges of the label switched network 110 may be configured to insert or remove the labels of the packets transported between the switched network 110 and external networks. The first node 112 and the last node 112 along a path are sometimes referred to as the source node and the destination node, respectively. Although four nodes 112 are shown in the label switched network 110, the label switched network 110 may comprise any quantity of nodes 112. Additionally, the nodes 112 may be located in different domains in the label switched network 110 and may be configured to communicate across multiple domains. For example, the nodes 112 that correspond to different domains may exchange packets along a path that may be established across multiple domains.
In an embodiment, the control plane controller 120 may be any device configured to coordinate activities within the label switched network 110, such as a Network Management System (NMS) or Operations Support System (OSS). Specifically, the control plane controller 120 may receive routing requests from the label switched network 110 and provide back the corresponding path information. In addition, the control plane controller 120 may communicate with the PCE 130, for instance using a PCE Protocol (PCEP), provide the PCE 130 with information that may be used for path computation, receive the computed path from the PCE 130, and forward the computed path to at least one of the nodes 112. The control plane controller 120 may be located in a component outside of the label switched network 110, such as an external server, or may be located in a component within the label switched network 110, such as a node 112.
In an embodiment, the PCE 130 may be any device configured to perform all or part of the path computation for the label switched network 110, e.g. based on a path computation request. Specifically, the PCE 130 may receive the information that may be used for computing the path from the control plane controller 120, from the node 112, or both. The PCE 130 may then process the information to obtain the path. For instance, the PCE 130 may compute the path, and determine the nodes 112 including the LSRs along the path. The PCE 130 may then send all or part of the computed path information to the control plane controller 120 or directly to at least one node 112. Further, the PCE 130 may be coupled to or comprise a traffic-engineering database (TED), a P2MP Path database (PDB), a P2P path database, an optical performance monitor (OPM), a physical layer constraint (PLC) information database, or combinations thereof, which may be used to compute the path. The PCE 130 may be located in a component outside of the label switched network 110, such as an external server, or may be located in a component within the label switched network 110, such as a node 112. In an embodiment, a plurality of PCEs 130, which may be associated to a plurality of domains in the label switched network 110, may perform a distributed path computation across the domains based on a path computation request for an inter-domain P2MP tree, as described in detail below.
In an embodiment, the path computation request may be sent to the PCE 130 by a PCC. The PCC may be any client application requesting a path computation to be performed by the PCE 130. The PCC may also be any network component that makes such a request, such as the control plane controller 120, or any node 112, such as a LSR. For instance, the PCC may request from the PCE a P2MP path or P2P path in a single domain or across multiple domains in the label switched network 110. Additionally, the PCC may send the PCE 130 at least some of the path required information.
In an embodiment, the packets transported between network nodes, such as the nodes 112, are referred to as label switched packets, and may comprise labels that may be used to switch the packets along the nodes of a computed path. A path computed or given and signaled by MPLS for transporting or routing the label switched packets may be referred to as a LSP. For example, the LSP may be a TE LSP established using a Resource Reservation Protocol-Traffic Engineering (RSVP-TE). The LSP may be a P2P TE LSP that extends from a source node to a destination node and may be unidirectional, where the packets may be transported in one direction along the path, e.g., from the source node to the destination node in the label switched network 110. Alternatively, the LSP may be a P2MP TE LSP, which may extend from a source or root node to a plurality of destination or leaf nodes. The P2MP TE LSP may be considered as a combination of a plurality of P2P TE LSPs that share the same source node. In some embodiments, the P2MP TE LSP is referred to as a P2MP tree and its P2P TE LSPs are referred to as Source-to-Leaf (S2L) sub-LSPs. The P2MP tree may be used to provide multicast services, such as multicast Virtual Private Networks (VPNs), Internet Protocol Television (IPTV), content-rich media distribution, other high-capacity applications, or combinations thereof. Further, the P2MP tree may be an inter-domain P2MP tree, where the source node and the leaf nodes may be distributed across multiple domains, e.g. in the label switched network 110.
Typically, an inter-domain P2MP tree may be computed using a distributed PCE architecture across multiple domains. For example, a plurality of PCEs may compute a plurality of branching paths in different domains from a source node in one of the domains. The branching paths may extend to the different domains that comprise the leaf nodes. As such, the P2MP path computation may result in a plurality of path options per domain, which may be difficult to coordinate efficiently between the domains. For instance, determining which of the border nodes in the different domains to use for connecting the paths across the domains may be difficult.
One approach to compute the inter-domain P2MP tree is to compute a plurality of shortest inter-domain P2P paths from the source node to each destination or leaf node, and then combine the paths to obtain the inter-domain P2MP tree, e.g. a Steiner P2MP tree. However, the Steiner P2MP tree computation may require the replication of incoming packets to each of the P2P paths at the ingress or source node to accommodate multipoint communications. Such requirement may increase the processing burden on the ingress node and may not be practical in terms of scalability. Additionally, the Steiner P2MP tree computation may not make use of bandwidth sharing when a plurality of P2P paths share at least one link along the paths, which may waste bandwidth resources, memory, MPLS label space, or combinations thereof. Further, the tree may require changing or substantial reconfiguration to add and/or remove a single destination or leaf node from the tree. Therefore, the frequency of the Steiner P2MP tree computation, e.g. in response to link and/or node updates, may be computationally intensive and the resulting reconfiguration may cause system destabilization.
The first domain 212 may comprise the source node 202, a branch node 204, a transit node 206, a leaf node 210, a bud node 211, and two BNs 208. The ingress or source node 202 may be coupled to a first network site 290 (site A) and the branch node 204, and may be configured to multicast traffic to the leaf nodes 210 via the path of the inter-domain P2MP tree 200. For example, the source node 202 may forward traffic from the site A, which may correspond to a service provider, to the branch node 204 along a link in the inter-domain P2MP tree 200. The branch node 204 may be coupled to the transit node 206, the bud node 211, and one of the BNs 208, and may be configured to multicast traffic from the source node 202 to the transit node 206, the bud node 211, and the BN 208 via a plurality of links in the tree. The transit node 206 may be coupled to another BN 208 and may be configured to forward the traffic from the branch node 204 to the BN 208 via a link in the tree.
The two BNs 208 may be individually coupled to a BN 208 in the second domain 220 and another BN 208 in the third domain 230 and may be configured to forward the traffic to the BNs 208 in the second domain 220 and the third domain 230 via corresponding links in the tree. The bud node 211 may be coupled to a third network site 290 (site B) and may be configured to forward the traffic from the branch node 204 to the site B. However, the bud node 211 may also be couple to the leaf node 210 and may forward traffic to the leaf node 210 via a link in the tree. The leaf node 210 may be coupled to a second network site 290 (site C) and may be configured to forward the traffic from the bud node 211 to the site C, which may correspond to a client.
The second domain 220 may comprise a transit node 206 and two BNs 208. The transit node 206 may be positioned between the two BNs 208 and may forward traffic between the two BNs 208 via links in the tree. The two BNs 208 may be coupled individually to a BN 208 in the first domain 212 and another BN 208 in the fourth domain 240, respectively, and may forward the multicast traffic from the first domain 212 to the fourth domain 240 via corresponding links in the tree. Similarly, the third domain 230 may comprise a transit node 206 and two BNs 208. The transit node 206 may be positioned between the two BNs 208 and may forward traffic between the two BNs 208 via links in the tree. The two BNs 208 may be coupled individually to a BN 208 in the first domain 212 and another BN 208 in the fourth domain 240, respectively, and may forward the multicast traffic from the first domain 212 to the fourth domain 240 via corresponding links in the tree.
The fourth domain 240 may comprise two BNs 208, two transit nodes 206, a branch node 204, and three leaf nodes 210. The two BNs 208 may be coupled individually to a BN 208 in the second domain 220 and another BN 208 in the third domain 230, respectively, and may forward the multicast traffic from the second domain 220 and the third domain 230 to the fourth domain 240 via corresponding links in the tree. Each of the two transit nodes 206 may be individually coupled to one of the two BNs 208. However, one of the transit nodes 206 may be coupled to the branch node 204, and the other transit node 206 may be coupled to a one of the leaf nodes 210. Thus, the two transit nodes 206 may individually forward the traffic from the two BNs 208 to the branch node 204 and the leaf node 210 via corresponding links in the tree. The branch node 204 may also be coupled to two other leaf nodes 210, and may multicast the traffic from one of the transit nodes 206 to the two leaf nodes 210. Each of the three leaf nodes 210 in the fourth domain 240 may be coupled to a different network site 290 (site D, site E, and site F), respectively, and may forward the traffic to the corresponding network site 290.
In an embodiment, the computation of the inter-domain P2MP tree 200 paths between the source node 202 and the leaf nodes 210 may be distributed between the first PCE 214, the second PCE 222, the third PCE 232, and/or the fourth PCE 242. For instance, the first PCE 214 may receive a request message from a PCC in the first domain 212, which may be the ingress or source node 202, to compute the inter-domain P2MP tree 200. The request message may indicate that the computation is related to an inter-domain P2MP path. Additionally, the message may comprise path computation information, which may be used to calculate the paths across the domains. The first PCE 214, the second PCE 222, the third PCE 232, and the fourth PCE 242 may communicate with each other to share the path computation information and perform a combined path computation. The PCEs may also share information during path computation to synchronize information and obtain an improved P2MP path computation.
In an embodiment, the inter-domain P2MP tree 200 computation may comprise determining (e.g. at the first PCE 214) which domains may be used to establish end-to-end paths between the source node 202 and each leaf node 210. The inter-domain P2MP tree 200 computation may also comprise selecting the BNs 208 to connect the paths across the domains, selecting the branch nodes 204 to branch out the paths in the tree, and computing a minimum cost tree across the domains according to some metric, such as minimum cost, shortest path, minimizing a load on a most used link, maximizing a load on a least used link, balancing the loads on a plurality of links, or any other metric. In some embodiments, the tree computation may also comprise computing a backup tree, which may comprise a plurality of different nodes, paths, and/or links that may be different from the components in the primary computed tree.
The P2MP path computation may be performed in separate procedures: a core tree computation that may be performed jointly by the PCEs, and a prior, concurrent, or subsequent sub-tree computation that may be performed individually by at least one of the PCEs. The PCEs may compute the core tree across the domains, which may extend from the source node 202 in the first domain 212 to the BNs 208 in each of the remaining domains. The core tree may be established by computing a plurality of inter-domain P2P paths across the domains. For instance, the core tree may be computed using a BRPC technique, e.g. without sharing domain private information between the domains. The sub-trees may be computed by at least some of the PCEs, which may be associated with the domains that comprise the leaf nodes 210, e.g. the first PCE 214 and the fourth PCE 242. The sub-trees may extend from the BN 208 to the leaf nodes 210 in the corresponding domains, e.g. the first domain 212 and the fourth domain 240. The core tree and the sub-tree may be computed based on a request message, which may comprise a bit that may be set, e.g. in a request object, to indicate a core tree computation or a sub-tree computation request.
The P2MP path computation may be configured to preserve confidentiality information across the domains, which may be required when the domains are managed by different service providers. For example, the sub-trees that comprise the leaf nodes 210 in each domain may be hidden from the other domains. Further, computing the core tree and the sub-trees separately, e.g. using separate procedures and/or techniques, may improve inter-domain P2MP tree computation in terms of efficiency, scalability, and/or resource sharing, e.g. in comparison to the Steiner P2MP tree computation approach.
In the first domain 312, the core tree may comprise the source node 302 (S) and a BN 308 (A) coupled to the source node 302. The first domain 312 may also comprise a first sub-tree that comprises two branch nodes 304 and three leaf nodes 310. In the first sub-tree, one of the branch nodes 304 may be coupled to the source node 302 (S), to one of the leaf nodes 310, and to another branch node 304, which may be coupled to the other two leaf nodes 310. In the second domain 320, the core tree may comprise two BNs 308 (E and M) and three transit nodes 306 (P, Q, and R) positioned between the BNs 308 E and M. The BN 308 E in the second domain 320 may also be coupled to the BN 308 A in the first domain 312. Additionally, the core tree may comprise two BNs 308 (W and T) in the third domain 330, and a BN 308 (D1) in the fourth domain 340. The BN 308 W in the third domain 330 may be coupled to the BN 308 M in the second domain 320. The fourth domain 340 may also comprise a second sub-tree that comprises two branch nodes 304 and four leaf nodes 310. In the second sub-tree, each branch node 304 may be coupled to two of the leaf nodes 310 and to the BN 308 D1 of the core tree, which may also be coupled to the BN 308 T in the third domain 330.
In the fifth domain 350, the core tree may comprise two BNs 308 (X and U) and a branch node 304 (Z), which may be positioned between the two BNs 308 X and U. The fifth domain 350 may also comprise a third sub-tree that comprises two branch nodes 304 and four leaf nodes 310. In the third sub-tree, each branch node 304 may be coupled to two of the leaf nodes 310 and to the branch node 304 Z of the core tree. The BN 308 X in the fifth domain 350 may also be coupled to the BN 308 M in the second domain 320. In the sixth domain 360, the core tree may comprise a BN 308 (D2) coupled to BN 308 U and a fourth sub-tree that comprises three branch nodes 304 and four leaf nodes 310. In the fourth sub-tree, one of the branch nodes 304 may be coupled to the BN 308 D2 of the core tree and to two other branch nodes 304, which may each be coupled to two leaf nodes 310.
The core tree in the first domain 312, the second domain 320, the third domain 330, the fourth domain 340, the fifth domain 350, and the sixth domain 360 may be computed by the first PCE 314, the second PCE 322, the third PCE 332, the fourth PCE 342, the fifth PCE 352, the sixth PCE 362, or combinations thereof. In a specific embodiment, the first sub-tree in the first domain 312 may be computed by the first PCE 314, the second sub-tree in the fourth domain 340 may be computed by the fourth PCE 342, the third sub-tree in the fifth domain 350 may be computed by the fifth PCE 352, and the fourth sub-tree in the sixth domain 360 may be computed by the sixth PCE 362.
The topology of the PCEs associated with the domains of the inter-domain P2MP tree 300 may be organized in a distributed PCE topology. The distributed PCE topology may comprise a root PCE, a transit PCE, a branch PCE, and a leaf PCE. The root PCE may be associated with the domain that comprises the source node 302, e.g. the first PCE 314. The transit PCE may be associated with the domain that transfers the traffic from the source node 302 to another domain, e.g. the third PCE 332. The branch PCE may be associated with the domains that forward the traffic from the source node 302 to a plurality of other domains, e.g. the second PCE 322. The leaf PCE may be associated with the domains that comprise a leaf node, e.g. the fourth PCE 342, the fifth PCE 352, and the sixth PCE 362. The distributed PCE topology may be communicated between the PCEs, e.g. during the discovery period of the PCEs or via a path computation request to each PCE.
As shown in
In an embodiment, the core tree may be computed by building the core tree BNs, e.g. selecting the BNs 308 of the core tree. As such, the quantity of BNs may be reduced or minimized to improve path computation efficiency. Building the core tree BNs may be achieved using a BRPC procedure and a Virtual Shortest Path Tree (VSPT) procedure. The VSPT procedure may comprise computing a tree, which may have an egress at the root of the core tree and an ingress at a leaf of the core tree. A primary PCE (e.g. the first PCE 314), which may receive a request from a PCC (e.g. the source node 302 S), may compute a plurality of potential core trees using the VSPT procedure, e.g. by communicating with other PCEs associated with multiple domains. The primary PCE may then determine an optimal core tree, e.g. according to cost and/or bandwidth requirements. Subsequently, a plurality of sub-trees may be built to connect the leaf nodes of the core tree, which may be BNs in the core tree, to the destination nodes (e.g. leaf nodes 310).
The paths 400 may comprise end-to-end paths between the source node 302 S in the first domain 312 and the BN 308 D1 in the fourth domain 340. The BN 308 D1 may be an entry (ingress edge) node to the fourth domain 340, which may be one of the domains that comprise destination or leaf nodes of the inter-domain P2MP tree 300. The end-to-end paths between the source node 302 S and the BN 308 D1 may be used to compute a branch or portion of the core tree in the inter-domain P2MP tree 300, which may be subsequently extended and linked to the leaf nodes 310 in the fourth domain 340 via a sub-tree, as described below. Similarly, the source node 302 S may also be extended and linked to the leaf nodes 310 in the first domain 312 via a sub-tree.
The first sub-path D1-1 may comprise the source node 302 S, the BN node 308 D1, and a plurality of boundary nodes 308 A, E, G, V, and T between the source node 302 S and the BN node 308 D1. The first sub-path D1-1 may also comprise the transit node 306 F (not shown) between the BNs 308 E and G, which may correspond to about two hops in the path. The second sub-path D1-2 may comprise the source node 302 S, the BN node 308 D1, and a plurality of boundary nodes 308 A, E, M, W, and T between the source node 302 S and the BN node 308 D1. The second sub-path D1-2 may also comprise the transit nodes 306 P, Q, and R (not shown) between the BNs 308 E and M, which may correspond to about four hops in the path. The third sub-path D1-3 may comprise the source node 302 S, the BN node 308 D1, the transit node 306 B, and a plurality of boundary nodes 308 C, H, E, G, V, and T between the source node 302 S and the BN node 308 D1. The third sub-path D1-3 may also comprise the transit node 306 P (not shown) between the BNs 308 H and E, which may correspond to about two hops in the path, and the transit node 306 F (not shown) between the BNs 308 E and G, which may also correspond to about two hops. The fourth sub-path D1-4 may comprise the source node 302 S, the BN node 308 D1, the transit node 306 B, and a plurality of boundary nodes 308 C, H, M, W, and T between the source node 302 S and the BN node 308 D1. The fourth sub-path D1-4 may also comprise the transit nodes 306 P, Q, and R (not shown) between the BNs 308 H and M, which may correspond to about four hops in the path.
The sub-paths 500 may be computed using the VSPT procedure. Specifically, the fourth PCE 342 may first compute or select the BN 308 D1 in the fourth domain 340 and send this information to the third PCE 332 and/or any of the other PCEs. The third PCE 332 may then use this information to compute a first path comprising the BNs 308 V, T, and D1, and a second path comprising the BNs 308 W, T, and D1. The third PCE 332 may send this information to the second PCE 322 and/or any of the other PCEs. The second PCE 322 may then use this information to update and/or compute a first path comprising the nodes E, F, G, V, T, and D1, a second path comprising the nodes E, P, Q, R, M, W, T, and D1, a third path comprising the nodes H, P, E, F, G, V, T, D1, and a fourth path comprising the nodes H, P, Q, R, M, W, T, and D1. The second PCE 322 may send this information to the first PCE 314 and/or any of the other PCEs. The first PCE 314 may then use this information to compute the first sub-path D1-1 comprising the nodes S, A, E, F, G, V, T, and D1, the second sub-path D1-2 comprising the nodes S, A, E, P, Q, R, M, W, T, and D1, the third sub-path D1-3 comprising the nodes S, B, C, H, P, E, F, G, V, T, and D1, and the fourth sub-path D1-4 comprising the nodes S, B, C, H, P, Q, R, M, W, T, and D1.
As such, the first sub-path D1-1 may comprise about seven hops (e.g. connections) between the nodes S and D1, the second sub-path D1-2 may comprise about nine hops between the nodes S and D1, the third sub-path D1-3 may comprise about ten hops between the nodes S and D1, and the fourth sub-path D1-4 may comprise about ten hops between the nodes S and D1. The sub-paths 500 may be received by the first PCE 314 and may be considered to compute the core tree in the inter-domain P2MP tree 300, where at least one of the sub-paths 500 may be selected as a branch or a portion of the core tree, as described below.
The paths 600 may comprise end-to-end paths between the source node 302 S in the first domain 312 and the BN 308 D2 in the sixth domain 360. The BN 308 D2 may be an entry (ingress edge) node to the sixth domain 360, which may be one of the domains that comprise destination or leaf nodes of the inter-domain P2MP tree 300. The end-to-end paths between the source node 302 S and the BN 308 D2 may be used to compute a branch or portion of the core tree in the inter-domain P2MP tree 300, which may be subsequently extended and linked to the leaf nodes 310 in the sixth domain 360 via a sub-tree, as described below. Similarly, the branch node 304 Z may also be extended and linked to the leaf nodes 310 in the fifth domain 350 via a sub-tree.
The first sub-path D2-1 may comprise the source node 302 S, the BN node 308 D2, and a plurality of boundary nodes 308 A, E, M, X, and U between the source node 302 S and the BN node 308 D2. The first sub-path D2-1 may also comprise the transit nodes 306 P, Q, and R (not shown) between the BNs 308 E and M, which may correspond to about four hops in the path. In
The sub-paths 700 may be computed using the VSPT procedure. Accordingly, the sixth PCE 362 may first compute or select the BN 308 D2 in the sixth domain 360 and send this information to the fifth PCE 352 and/or any of the other PCEs. The fifth PCE 352 may then use this information to compute a first path comprising the nodes X, Z, U, and D2, and a second path comprising the BNs 308 Y, U, and D2. The fifth PCE 352 may send this information to the second PCE 322 and/or any of the other PCEs. The second PCE 322 may then use this information to compute the first sub-path comprising the nodes E, P, Q, R, M, X, Z, U and D2, the second sub-path comprising the nodes E, P, H, J, K, Y, U, and D2, the third sub-path comprising the nodes H, P, Q, R, M, X, Z, U, and D2, and the fourth sub-path comprising the nodes H, J, K, Y, U, and D2. The second PCE 322 may send this information to the first PCE 314 and/or any of the other PCEs. The first PCE 314 may then use this information to compute the first sub-path D2-1 comprising the nodes S, A, E, P, Q, R, M, X, Z, U and D2, the second sub-path D2-2 comprising the nodes S, A, E, P, H, J, K, Y, U and D2, the third sub-path D2-3 comprising the nodes S, B, C, H, P, Q, R, M, X, Z, U, and D2, and the fourth sub-path D2-4 comprising the nodes S, B, C, H, J, K, Y, U and D2.
As such, the first sub-path D2-1 may comprise about ten hops between the nodes S and D2, the second sub-path D2-2 may comprise about nine hops between the nodes S and D2, the third sub-path D3-3 may comprise about 11 hops between the nodes S and D2, and the fourth sub-path D4-4 may comprise about eight hops between the nodes S and D2. The sub-paths 700 may be considered by the first PCE 314 to compute the core tree in the inter-domain P2MP tree 300, where at least one of the sub-paths 700 may be selected as a branch or a portion of the core tree, as described below.
For instance, at least a portion of any of the sub-paths 700 may be merged with at least a portion of any of the sub-paths 400 to obtain at least a portion of the core tree in the inter-domain P2MP tree 300. For example, the sub-paths may be merged to obtain end-to-end paths from the source node 302 S to the leaf nodes 310 that have a minimum number of hops and hence minimum cost. In an embodiment, an algorithm may be used to merge the sub-paths computed using the VSPT procedure and hence obtain a core tree. According to the algorithm, each computed sub-path may be merged with another computed sub-path, and the cost of the resulting merged sub-path may be calculated. For example a sub-path D1-i from the sub-paths 500 may be merged with a sub-path D2-j from the sub-paths 700, where i and j are integer values that may be changed to select all the sub-paths 500 and 700 that may be merged. The process may be repeated until the costs of all possible merged sub-paths are calculated. Finally, the merged sub-path that has the least cost may be selected to establish the core tree. In an embodiment, the cost of each merged sub-path may be equal to the quantity of hops of the merged sub-path. However, in other embodiments, other criteria may be used to evaluate the requirements of each path, such as bandwidth and/or quality of service (QoS) requirements.
In an embodiment, the algorithm for merging the sub-paths 500 and 700 and obtaining the core tree may comprise the following instructions:
1—Take a sub-path(D1-i) from VSPT(D1) and a sub-path(D2-j) from VSPT(D2).
2—Merge the sub-path(D1-i) and the sub-path(D2-j) into a Core Tree(D1-i-D2-j).
3—Compute the cost of the Core Tree(D1-i-D2-j).
4—Repeat (1) to (3) for all i and j combinations to generate all the possible Core Trees.
5—Evaluate and identify the Core Tree with the minimum cost among all the Core Trees.
In an embodiment, to complete the inter-domain P2MP tree computation, a plurality of sub-trees that comprise the destination or leaf nodes may be computed and then combined with a core tree. The sub-tree may be computed in each domain that comprises the leaf nodes, e.g. independent of the sequence of domains with respect to the core tree and/or the inter-domain P2MP tree. The sub-tree may be computed and optimized, e.g. based on an objective function (OF), which may comprise at least one optimization criterion that may be used to compute a path (e.g. path cost minimization) or to synchronize computation of a plurality of paths (e.g. aggregate bandwidth consumption minimization). For example, the optimized sub-tree may be a minimum cost tree or a shortest path tree. Further, the sub-tree computation in one domain may be independent of the computation of another sub-tree in another domain.
In an embodiment, the PCC (e.g. the source or ingress node) and PCEs may exchange a request message and reply message to compute new paths, to add branches to existing paths, or to store, delete, or re-optimize paths. The quantity of exchanged request and/or reply messages may be independent of the quantity of requested leaf nodes in the tree. The messages exchanged between the PCC and PCEs may indicate whether the computation request or reply is related to a P2MP path or P2P path. Additionally, the messages may comprise path computation information, which may be used to request or compute the path. For instance, the messages may comprise a request/reply (RP) object that indicates a standard P2MP path (e.g. in a single domain) or an inter-domain P2MP path related message, a PCE sequence object that specifies the distributed PCE topology associated with the domains, and/or an end-point object that specifies a source and at least one destination node for the path. The messages may also comprise an error object that indicates a computation failure, at least some nodes in the request message that may not be used, or both.
In an embodiment, the C flag may be set to indicate that the request or reply message is related to an inter-domain P2MP path or tree computation. Further, at least some of the fields of the RP object 1300 may be configured based on the PCEP. For instance, the Reserved field 1321 may be reserved for other purposes and/or may not be used. The O flag 1323 may be set in a request message to indicate that a loose path is acceptable or may be cleared to indicate that a path comprising exclusively strict hops is required. On the other hand, the O flag 1323 may be set in a reply message to indicate that the computed path is loose or may be cleared to indicate that the computed path comprises strict hops. The B flag 1324 may be set to indicate that a path computation request relates to at least one bidirectional TE LSP, which may have the same TE requirements in each direction, such as fate sharing, protection and restoration, LSRs, TE Links, resource requirements (e.g., latency and jitter), etc. Otherwise, the B flag 1324 may be cleared to indicate that the LSP is unidirectional. The R flag 1325 may be set to indicate that a computation request relates to re-optimizing an existing path or branch. The Pri flags 1326 may be used to specify a recommended request priority. For instance, the Pri flags 1326 may have a value from about one to about seven, which may be set locally at the PCC. Alternatively, the P flags 1326 may be set to zero when the request priority is not specified. The Request-ID-number 1330 may be combined with the source IP address of the PCC or the PCE network address to identify the path computation request context. The Request-ID-number may be changed or incremented each time a new request is sent to the PCE. In some embodiments, the E flag may be set to indicate that the path information is represented in a compressed format or may be cleared otherwise. The M flag may be set to indicate whether the request message or reply message is related to a P2MP path or P2P path computation.
In an embodiment, the inter-domain P2MP path computation messages may be configured to satisfy at least some of the following requirements:
1—Indication of P2MP Path computation request.
2—Indication of P2MP objective functions.
3—Non-support of P2MP path computation.
4—Non-support by back-level PCE implementations.
5—Specification of destinations.
6—Indication of P2MP paths.
7—Multi-message requests and responses.
8—Non-specification of per-destination constraints and parameters.
9—Path modification and path diversity.
10—Re-optimization of P2MP TE LSPs.
11—Addition and removal of destinations from existing paths.
12—Specification of applicable branch nodes.
13—Capabilities exchange.
14—Multi-message request and response.
In an embodiment, the PCC and PCEs may exchange a reply message, which may have the following format:
In an embodiment, the PCC and PCEs may exchange a request message, which may have the following format:
In an embodiment, a plurality of manageability requirements described in the Internet Engineering Task Force (IETF) document draft-yasukawa-pce-p2mp-req-05 and the IETF Request for Comments (RFC) 5440, which are incorporated herein by reference as if reproduced in their entirety, may be used to support inter-domain P2MP path or tree computation. For instance, the PCE may require information about the P2MP signaling and branching capabilities of the nodes (e.g. LSRs) in the network. The PCEP protocol may also be extended to satisfy additional manageability requirements. For instance, a plurality of control parameters may be used and/or exchanged, e.g. for function and/or policy control. The control parameters may comprise a parameter to enable or disable P2MP path computation, enable or disable P2MP path computation capability advertisement (e.g. using discovery protocol or capability exchange), or both. Additionally, a plurality of management information base (MIB) objects, e.g. for information and data models, may be used to support inter-domain P2MP tree computation, as described in the IETF document draft-yasukawa-pce-p2mp-req-05.
The network components described above may be implemented on any general-purpose network component, such as a computer or network component with sufficient processing power, memory resources, and network throughput capability to handle the necessary workload placed upon it.
The secondary storage 1604 is typically comprised of one or more disk drives or erasable programmable ROM (EPROM) and is used for non-volatile storage of data. Secondary storage 1604 may be used to store programs that are loaded into RAM 1608 when such programs are selected for execution. The ROM 1606 is used to store instructions and perhaps data that are read during program execution. ROM 1606 is a non-volatile memory device that typically has a small memory capacity relative to the larger memory capacity of secondary storage 1604. The RAM 1608 is used to store volatile data and perhaps to store instructions. Access to both ROM 1606 and RAM 1608 is typically faster than to secondary storage 1604.
At least one embodiment is disclosed and variations, combinations, and/or modifications of the embodiment(s) and/or features of the embodiment(s) made by a person having ordinary skill in the art are within the scope of the disclosure. Alternative embodiments that result from combining, integrating, and/or omitting features of the embodiment(s) are also within the scope of the disclosure. Where numerical ranges or limitations are expressly stated, such express ranges or limitations should be understood to include iterative ranges or limitations of like magnitude falling within the expressly stated ranges or limitations (e.g., from about 1 to about 10 includes, 2, 3, 4, etc.; greater than 0.10 includes 0.11, 0.12, 0.13, etc.). For example, whenever a numerical range with a lower limit, Rl, and an upper limit, Ru, is disclosed, any number falling within the range is specifically disclosed. In particular, the following numbers within the range are specifically disclosed: R=Rl+k*(Ru−Rl), wherein k is a variable ranging from 1 percent to 100 percent with a 1 percent increment, i.e., k is 1 percent, 2 percent, 3 percent, 4 percent, 5 percent, . . . , 50 percent, 51 percent, 52 percent, . . . , 95 percent, 96 percent, 97 percent, 98 percent, 99 percent, or 100 percent. Moreover, any numerical range defined by two R numbers as defined in the above is also specifically disclosed. Use of the term “optionally” with respect to any element of a claim means that the element is required, or alternatively, the element is not required, both alternatives being within the scope of the claim. Use of broader terms such as comprises, includes, and having should be understood to provide support for narrower terms such as consisting of, consisting essentially of, and comprised substantially of. Accordingly, the scope of protection is not limited by the description set out above but is defined by the claims that follow, that scope including all equivalents of the subject matter of the claims. Each and every claim is incorporated as further disclosure into the specification and the claims are embodiment(s) of the present disclosure. The discussion of a reference in the disclosure is not an admission that it is prior art, especially any reference that has a publication date after the priority date of this application. The disclosure of all patents, patent applications, and publications cited in the disclosure are hereby incorporated by reference, to the extent that they provide exemplary, procedural, or other details supplementary to the disclosure.
While several embodiments have been provided in the present disclosure, it should be understood that the disclosed systems and methods might be embodied in many other specific forms without departing from the spirit or scope of the present disclosure. The present examples are to be considered as illustrative and not restrictive, and the intention is not to be limited to the details given herein. For example, the various elements or components may be combined or integrated in another system or certain features may be omitted, or not implemented.
In addition, techniques, systems, subsystems, and methods described and illustrated in the various embodiments as discrete or separate may be combined or integrated with other systems, modules, techniques, or methods without departing from the scope of the present disclosure. Other items shown or discussed as coupled or directly coupled or communicating with each other may be indirectly coupled or communicating through some interface, device, or intermediate component whether electrically, mechanically, or otherwise. Other examples of changes, substitutions, and alterations are ascertainable by one skilled in the art and could be made without departing from the spirit and scope disclosed herein.
The present application claims priority to U.S. Provisional Patent Application No. 61/153,769 filed Feb. 19, 2009 by Qianglin Quintin Zhao et al. and entitled “System and Method for Point to Multipoint Inter-Domain MPLS Traffic Engineering Path Calculation,” which is incorporated herein by reference as if reproduced in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
7406481 | Saha et al. | Jul 2008 | B2 |
7701940 | Novello et al. | Apr 2010 | B2 |
7801137 | Vasseur et al. | Sep 2010 | B2 |
7821951 | Douville | Oct 2010 | B2 |
20040184441 | Wu et al. | Sep 2004 | A1 |
20090245253 | Chen et al. | Oct 2009 | A1 |
Number | Date | Country |
---|---|---|
2007-074308 | Mar 2007 | JP |
2007-228087 | Sep 2007 | JP |
2008-206095 | Sep 2008 | JP |
2008-211656 | Sep 2008 | JP |
Number | Date | Country | |
---|---|---|---|
20100208733 A1 | Aug 2010 | US |
Number | Date | Country | |
---|---|---|---|
61153769 | Feb 2009 | US |