Scope abstraction extensions for redirection protocols

Abstract

Routing protocol scope abstraction system, encompassing the following: a network value comprising network devices, wherein the network comprises a first network area comprising first network devices and a second network area comprising second network devices; a selected area manager trained to distribute an area representation node link state packet (LSP) to inject link information regarding an area representation node representing the first network area to an area link state database of the second network devices in the second network area.

Description

BACKGROUND

[0001] Networks of interconnected devices (e.g., computer networks) often comprise any number of network devices. Such network devices can be arranged in a network topology to provide connectivity between devices inside and outside the network. Within such network topologies, routing protocols are often implemented that distribute specific information (e.g., network topology information) concerning the routing of network traffic data units (e.g., packets, frames, etc.) within and / or through the network. This information can be stored in corresponding databases (e.g., connection state databases) on the network devices.Including information describing the entire network topology or a substantial portion thereof in such databases may decrease the performance of the network or any devices within it, while increasing the amount of stored routing-related information.

[0002] Reference is made to “Dynamic Flooding on Dense Graphs; draft-li-dynamic-flooding-04.txt” by T. Li, Arista Networks, Internet Engineering Task Force (IETF), No. 4, March 27, 2018, pp. 1–17. The author points out that routing with link-state protocols in dense network topologies can lead to suboptimal convergence times due to the overhead associated with flooding. This can be addressed by reducing the density of the flooding topology. This document discusses the problem in detail and presents an architectural solution. Specific protocol changes for IS-IS, OSPFv2, and OSPFv3 are described in this document.

[0003] Reference is also made to US 7,177,951 B1, which discloses methods and apparatus for address management for a node that acts as a peer group leader for a peer group of nodes at one level of the hierarchy of a PNNI hierarchical network, wherein the peer group leader represents the peer group to one or more neighboring nodes at the next higher level of the hierarchy. The peer group leader has a memory for storing peer group topology data, which includes address data provided to the peer group leader by nodes in the peer group and represents addresses for access by the network, as well as peer group topology data, which includes address data provided to neighboring nodes by the peer group leader and represents addresses that can be accessed via the peer group.Address management procedures, comprising: checking whether addresses represented by the address data are accessible via the peer group; notifying neighboring nodes of changes in the accessibility of the addresses thus identified; and updating the peer group leader topology data according to the changes. SUMMARY

[0004] According to a first aspect of the invention, a method for routing protocol area abstraction according to claim 1 is provided.

[0005] According to a second aspect of the invention, a non-transitory computer-readable medium is provided according to claim 7.

[0006] According to a third aspect of the invention, a system for routing protocol area abstraction according to claim 8 is provided. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 shows a network topology according to one or more embodiments of the invention. Fig. Figure 2 shows a system according to one or more embodiments of the invention. Fig. Figure 3 shows a flowchart according to one or more embodiments of the invention. Fig. Figure 4 shows an example according to one or more embodiments of the invention. DETAILED DESCRIPTION

[0007] Specific embodiments are now described with reference to the accompanying figures. Numerous details are listed in the following description as examples of the invention. Those skilled in the art, who have the advantage of this detailed description, will understand that one or more embodiments of the present invention can be carried out without these specific details and that numerous variations or modifications are possible without departing from the scope of the invention. Certain details known to those skilled in the art may be omitted to avoid complicating the description.

[0008] In the following description of the figures, each component described in relation to a figure in various embodiments of the invention may correspond to one or more components of the same name shown and / or described in relation to another figure. For the sake of brevity, descriptions of these components are not repeated in relation to each figure. Thus, each embodiment of the components of each figure is included by reference and is assumed to be optionally present in any other figure that has one or more components of the same name. Furthermore, according to various embodiments of the invention, each description of the components of a figure is to be interpreted as an optional embodiment that may be implemented in addition to, in conjunction with, or instead of the embodiments described in relation to a corresponding component of the same name in another figure.

[0009] Throughout the application, ordinal numbers (e.g., first, second, third, etc.) may be used as adjectives for an element (i.e., any noun in the application). The use of ordinal numbers is not intended to imply or create a particular order of elements, nor to restrict an element to a single element unless expressly disclosed, for example, by the use of terms such as "before," "after," "only," and other such terms. Rather, the use of ordinal numbers serves to distinguish between elements. For example, a first element is different from a second element, and the first element may comprise more than one element and follow (or precede) the second element in a sequence of elements.

[0010] In general, embodiments of the invention relate to the abstraction of an area within a network that implements a link-state routing protocol. In particular, in one or more embodiments of the invention, an area is abstracted to provide routing using network devices within a first area, without requiring the entire topology of the first area to be contained in link-state databases of network devices in a second area that is operationally connected to one or more network devices in the first area. Specifically, in one or more embodiments of the invention, a first area is abstracted as an area representation node adjacent to network devices in a second area.

[0011] In one or more embodiments of the invention, the network devices of the second domain report a proximity to the domain representation node instead of to specific network devices within the topology of the first domain, which is represented by the domain representation node. Thus, according to the connection state databases of network devices in the second domain, it may appear that a network traffic data unit (e.g., frame, packet, etc.) received at a first network device in the second domain, which is to be routed to a second network device in the second domain, can simply be routed to the second network device via the domain representation node.In this way, the actual topology of the first area and the network devices therein do not need to be included in the connection state databases of the network devices in the second area, which can reduce the memory load in such connection state databases. In one or more embodiments of the invention, a network traffic data unit can actually be routed through devices in the first area (i.e., the area represented by the area representation node) through area boundary devices (i.e., network devices that are operationally connected to both the first and the second area) using segment routing.

[0012] Although the various embodiments of the invention discussed below are described in connection with a first and a second region of a network, a person skilled in the art will recognize that embodiments of the invention are applicable to networks with any number of regions (e.g., more than two regions) and that in such networks, each of one or more regions can be represented as region representation nodes in each connection state database of other regions (e.g., there can be a hierarchy of region abstractions). For example, there can be a third region that encompasses both the first and the second region, with the internal topology details of the first region not exposed to the connection state database of the third region. In other words, the region abstraction can hide topology details from both adjacent and parent regions.

[0013] Fig. Figure 1 shows a network topology according to one or more embodiments of the invention. As in Fig. As shown in Figure 1, the network topology (100) comprises a backbone layer (102) and a leaf layer (110). The backbone layer (102) can include any number of backbone devices, such as backbone device A (104), backbone device B (106), and backbone device N (108). The leaf layer (110) can include any number of leaf devices, such as leaf device A (112), leaf device B (114), and leaf device N (116). Each of these components is described below.

[0014] In one or more embodiments of the invention, a network topology (100) is an arrangement of various network elements. In one or more embodiments of the invention, a network comprises a collection of one or more network devices (e.g., backbone devices (104, 106, 108), leaf devices (112, 114, 116)) that facilitate network connectivity for one or more operationally connected devices (e.g., computing devices, data storage devices, other network devices, etc.). As used herein, the term operationally connected or operational link means that there is a direct or indirect connection between elements / components that allows the elements to interact with each other in some way. For example, such elements can exchange information, send instructions to perform actions, cause changes in state and / or operating state, etc.Additionally, as used herein, a network can be an entire network or any part thereof (e.g., a logical part of network devices within the network topology, outside the network topology, etc.). A network can include a data center network, a wide area network, a local area network, a wireless network, a cellular network, or any other suitable network that facilitates the exchange of information from one part of the network to another. In one or more embodiments, the network may be connected to the Internet or at least partially overlap with it.

[0015] In one or more embodiments of the invention, the network topology (100) is a bipartite network topology. In one or more embodiments of the invention, a bipartite network topology is a network topology in which there are at least two sets of network devices, with connections between devices of one set and devices of another set, but no connections between devices within a given set. Examples of such network topologies include, but are not limited to, a folded three-level Clos network topology (e.g., a backbone-leaf network topology), a fat-tree topology (i.e., a Leiserson topology), etc.In one or more embodiments of the invention, a network topology can comprise network devices that are highly interconnected, wherein, for example, each device in a first set of network devices is connected to each device in a second set of network devices within the network topology. In the exemplary embodiment described in . Fig. As shown in Figure 1, the network topology (100) is a backbone-leaf topology. Although Fig. Figure 1 shows an example of a backbone-leaf topology. The network topology in one or more embodiments of the invention is not limited to being a backbone-leaf topology (or the particular example shown) or any other example of a network topology listed herein.

[0016] In one or more embodiments of the invention, the network topology (100), when arranged as a backbone-leaf topology, comprises a leaf layer (110) and a backbone layer (102).

[0017] In one or more embodiments of the invention, a sheet layer (110) is a set of any number of network devices (e.g., sheet devices (112, 114, 116)) that provide network connectivity to any number of attached devices (not shown), such as computing devices. In one or more embodiments of the invention, a computing device is a device or set of devices that can process instructions electronically and that includes at least one or more processors, memory, input and output devices, and a network connection via a sheet layer (110). Examples of computing devices include, but are not limited to, a server (e.g., a blade server in a blade server enclosure, a shelf server in a shelf, etc.), a virtual machine (VM), a desktop computer, a mobile device (e.g., a smartphone ...Laptop computers, smartphones, personal digital assistants, tablet computers and / or any other mobile computing device) and / or any other type of computing device meeting the aforementioned requirements. Other examples of devices for which a sheet layer (110) can provide network connectivity include, but are not limited to, data storage devices (not shown), other network devices (e.g., wireless routers) (not shown), media devices (e.g., televisions) (not shown), etc.

[0018] In one or more embodiments of the invention, a backbone layer (102) is a set of any number of network devices (e.g., backbone devices (104, 106, 108)) that provide at least network connectivity between network devices of a sheet layer (110). In one or more embodiments of the invention, each sheet device (112, 114, 116) within a sheet layer (102) is operationally connected to each backbone device (104, 106, 108) within a backbone layer (102), and each backbone device (104, 106, 108) within a backbone layer is functionally connected to each sheet device (112, 114, 116) within a sheet layer (102).

[0019] In one or more embodiments of the invention, each blade device (112, 114, 116) and each backbone device (104, 106, 108) is a network device. In one or more embodiments of the invention, a network device can be a physical device comprising and / or operationally connected to persistent memory (not shown), memory (e.g., random access memory (RAM)) (not shown), one or more processors (e.g., integrated circuits) (not shown), and two or more physical network interfaces (which may also be referred to as ports).

[0020] As in Fig. As shown in Figure 1, examples of such interfaces include ports S:1, S:2, and S:3 on each backbone device (104, 106, 108) and ports L:1, L:2, and L:3 on each blade device (112, 114, 116). Although in Fig. Not shown in Figure 1, blade devices (112, 114, 116) and / or backbone devices (104, 106, 108) can have any number of additional ports for connecting to any number of other devices. In one or more embodiments of the invention, the one or more processors of a network device (e.g., a central processing unit) are separate components of a network chip, one or more of which can also be components of a network device.

[0021] In one or more embodiments of the invention, the network device also comprises any number of network chips. In one or more embodiments of the invention, a network chip is any hardware (e.g., circuitry), software, firmware, and / or combination thereof that includes functionality for receiving, processing, and / or transmitting network traffic data units according to one or more embodiments of the invention. To perform such functionality, a network chip may comprise any number of components. Such components may, but are not limited to, include one or more processors, one or more buffers (e.g., for implementing receive and / or transmit queues, such as virtual output queues (VOQs)), any type or amount of non-volatile memory, and / or any type or amount of volatile memory (e.g., RAM).A network chip can also have any number of physical network interfaces (e.g., L:1, L:2, L:3, S:1, S:2 or S:3, as in . Fig. (1 shown) comprise a network device and / or be operationally connected to it. Such interfaces can provide a path outside the network device (e.g., to other devices) and / or be operationally connected to other components within the network device (100), and each such interface can be an input and / or output interface.

[0022] In one or more embodiments of the invention, a network device comprises a functionality for receiving network traffic data units (e.g., frames, packets, etc.) at any of the physical network interfaces (i.e., ports) of the network device and for processing the network traffic data units to determine whether: (i) the network traffic data unit is discarded; (ii) the network traffic data unit is processed; and / or (iii) the network traffic data unit is transmitted from a physical network interface or port on the network device based on the processing according to one or more embodiments of the invention.

[0023] As a non-restrictive example, a network chip can be hardware that receives network traffic data units at an input port and determines to which output port on the network device (100) the network traffic data units, such as Media Access Control Frames (MAC frames), which may contain Internet Protocol packets (IP packets), should be forwarded.

[0024] In one or more embodiments of the invention, a network device (e.g., leaf devices (112, 114, 116), backbone devices (104, 106, 108)) can include functionality for storing (e.g., in persistent memory, in storage, etc.) any number of data structures to facilitate the operation of at least some aspects of the network device. An example of such a data structure is a connection state database (not shown). Each network device described herein can include any number of connection state databases.

[0025] In one or more embodiments of the invention, a connection state database is a data store for storing connection state information received from neighboring network devices via the ports of a network device. In one or more embodiments of the invention, a data store is any type of storage unit(s) and / or device(s) (e.g., a file system, a database, a collection of tables, or any other storage mechanism) for storing data. Furthermore, the data store may comprise several different storage units and / or devices. The several different storage units and / or devices may be of the same type or not, or may be located in the same physical location.In one or more embodiments of the invention, the network device data storage (110) comprises all or any part of the persistent and / or non-persistent storage of the network device (100), as described above.

[0026] In one or more embodiments of the invention, network devices within a network topology (100) (or a section thereof (e.g., a logical area within the topology)) share link state information using a variety of data units (e.g., packets, frames, messages, indicators, etc.) which may be referred to herein as link state packets (LSPs). Such LSPs may be sent from a network device to directly connected adjacent network devices and may include information relating to the state of links between the sending network device and other devices with which the sending network device is operationally connected.

[0027] When a Link State Protocol (LSP) containing link state information is received at a network device from an adjacent network device, the receiving network device can store the information in at least one of its link state databases and propagate the LSP to its own adjacent network devices. Such sharing of link state information between network devices can occur within a network topology that implements an interior gateway protocol (IGP). Examples of IGPs include, but are not limited to, Inter-element to Inter-element (IS-IS) and Open Shortest Path First (OSPF), each of which can be considered a type of IGP known as a link-state routing protocol.

[0028] In one or more embodiments of the invention, when a link-state routing protocol is implemented within a given network topology (or any section thereof), each device participating in the link-state routing topology directly or indirectly receives link-state information from other devices in the network topology or a logical space within the network topology. In one or more embodiments of the invention, each network device uses the received information to establish a link-state compatibility database for network devices within the topology and / or logical space. Information relating to this compatibility database can be stored in a link-state database.

[0029] A network device with a complete mapping of the connectivity of network devices within a topology or logical space can then use any scheme (e.g., Dijkstra's algorithm, segment routing, etc.) to determine a path (e.g., shortest path, path with the lowest cost, etc.) from itself to other network devices in the network topology or logical space, and such information can be stored in the connection state database and / or used to program other sections of the network device (e.g., a routing information base) with information to be used, for example, when processing a received network traffic data unit.In one or more embodiments of the invention, the connectivity map on each network device in a topology or logical space should be the same, unless one or more network devices have a change in information relating to the state of one or more connections. Accordingly, once a mapping has been established, connection state information from a given network device can only be sent if a change in the connection state information of a connection of a network device occurs.

[0030] A non-restrictive example of a scheme for determining a path through a logical space can be called segment routing. In one or more embodiments of the invention, segment routing is a form of source routing in which a sending (i.e., source) device determines the route that a network traffic data unit takes through a network (or logical space) to a destination device. To achieve forwarding to the destination, a network device implementing segment routing can encapsulate a received network traffic data unit and any number of headers (e.g., IP packet headers, MAC frame headers, VXLAN headers, etc.) using any number of additional headers (e.g., labels) that tell a next-hop device what to do upon receiving the encapsulated network traffic data unit. For example, a first next hop (i.e.,The first next hop is instructed to send the encapsulated network traffic data unit to a second next hop. Accordingly, the first next hop can remove (e.g., subtract) the label containing instructions for the first next hop and then forward the remainder of the encapsulated network traffic data unit to the second next hop device. The process can be repeated until all added segment labels have been removed, and at which point the network traffic data unit can be processed by the destination device (e.g., because it has been received from a final destination, because it must further be forwarded toward a final destination device, etc.).

[0031] As explained above, link-state routing protocols can organize a network into logical scopes. For example, in the IS-IS protocol, a network can be organized into a Layer 1 (L1) scope used for intranet routing and a Layer 2 (L2) scope used for internet routing. Additionally, a network implementing IS-IS can include any number of network devices connected to both L1 and L2 devices. In other words, some devices can be described as participating in both L1 and L2, and thus be L1 / L2 devices. Devices participating in more than one logical scope (in any link-state routing protocol) can be referred to here as scope boundary devices.In one or more embodiments of the invention, devices in a backbone layer (102) can be grouped into a first region (e.g., an IS-IS-level-1 region), and devices in a sheet layer (110) can be grouped into a second region (e.g., an IS-IS-level-2 region), wherein any number of region-edge devices participate in both the first region and the second region, each of which can be either a sheet or a backbone device. Network regions and region-edge devices are further described in the description of . Fig. 2 discussed below.

[0032] In one or more embodiments of the invention, a network device also comprises software and / or firmware stored in any network device memory (not shown) and / or network device storage (not shown) (i.e., non-transitory computer-readable media). Such software may include instructions which, when executed by the one or more processors (not shown) of the network device, cause the one or more processors to perform actions according to one or more embodiments of the invention.The software instructions can be in the form of computer-readable program code to implement embodiments of the invention and can be stored wholly or partially, temporarily or permanently, on a non-transitory computer-readable medium such as a CD, DVD, storage device, floppy disk, tape, flash memory, physical memory, or other computer-readable storage medium. In particular, the software instructions can correspond to computer-readable program code which, when executed by one or more processors, is configured to perform functionality relating to embodiments of the invention. The functionality of a network device is not limited to the aforementioned examples.

[0033] Examples of a network device include, but are not limited to, a network switch, a router, a multilayer switch, a Fibre Channel device, an InfiniBand® device, etc. A network device is not limited to the aforementioned specific examples.

[0034] While Fig. Figure 1 shows a configuration of components; other configurations can be used without deviating from the scope of the invention. For example, the network topology can be any network topology other than the one shown. Fig. The backbone-leaf topology shown in Figure 1 is one example. As another example, the number of network devices in the backbone layer can be less (or greater) than the number of network devices in the leaf layer. As yet another example, each leaf device and each backbone device can have any number of additional ports for connecting to any number of other devices, or may not be connected to any other device at any given time. Accordingly, the embodiments disclosed herein should not be limited to the configuration of the devices shown in Figure 1. Fig. The components shown may be limited to 1.

[0035] Fig. Figure 2 shows a network implementing a link state protocol that divides the network into two areas according to one or more embodiments of the invention. As shown in Fig. As shown in Figure 2, the network comprises a first area (200), a second area (202), and any number of first-area / second-area (202) devices. Such first-area / second-area (202) devices may include, but are not limited to, an area conductor (204) and any number of area edge devices, such as area edge device A (206) and area edge device N (208). Additionally, the network may include any number of external devices, such as external device A (212) and external device N (214), and / or be operationally connected to them via the second area. Each of these components is described below.

[0036] In one or more embodiments of the invention, the first area (200) is a logical area of ​​a network topology comprising any number of interconnected network devices (not shown) to facilitate topology routing within the network. For example, the first area (200) can be the L1 area of ​​a network topology implementing the IS-IS link-state routing protocol. As another example, the first area can be the backbone area (i.e., area 0) of a network topology implementing the OSPF link-state routing protocol. In one or more embodiments of the invention, the network devices in the first area (200) are network devices located in a backbone layer of a network topology (e.g., backbone layer (102) of Fig. 1) are included, and as such, they can provide network connectivity between any number of network devices contained in a second area (210) (discussed below). In one or more embodiments of the invention, network devices in the first area (200) are operationally connected to at least any number of devices in a second area (210) and any number of devices that participate in both the first area (200) and the second area (210), such as devices of the first area / second area (202) (discussed below).

[0037] In one or more embodiments of the invention, the second area (210) is a logical area of ​​a network topology comprising any number of interconnected network devices (not shown) to facilitate the routing of network traffic data units that can be received by and / or sent to devices located outside the second area and / or the network topology comprising the first area (200) and the second area (210). For example, the second area (210) can be the L2 area of ​​a network topology implementing the IS-IS Link State Routing Protocol. As another example, the first area can be any non-backbone area (i.e., any area other than area 0) of a network topology implementing the OSPF Link State Routing Protocol.

[0038] In one or more embodiments of the invention, the network devices in the second area (210) are network devices that are in a layer of a network topology (e.g., backbone layer (110) of Fig. 1) are included, and as such, they can provide network connectivity for all connected devices (e.g., computing devices, other network devices, etc.). In one or more embodiments of the invention, network devices in the second area (210) are operationally connected to at least any number of devices in a first area (200) and any number of devices that participate in both the first area (200) and the second area (210), such as devices of the first area / second area (202) (discussed below).

[0039] In one or more embodiments of the invention, any number of external devices (e.g., external device A (212), external device N (214)) can be functionally connected to the second area. In one or more embodiments of the invention, an external device is any device that can be considered external to the network topology comprising the first area (200) and the second area (210). For example, an external device (212, 214) can be a shelf server directly connected to a top-of-rack switch located in the second area (210). As another example, an external device can be a top-of-rack switch connected to another network device located within the second area (210).As another example, an external device (212, 214) can be a mobile computing device that is functionally connected to a network device in the second area (210) via at least part of the Internet.

[0040] In one or more embodiments of the invention, any number of devices can be devices of the first area / second area (202). As in Fig. As shown in Figure 2, the dashed box labeled "first-area / second-area devices (202)" cannot actually be a separate area within a network topology implementing a link-state routing protocol, but is included merely to show that such devices can exist. In one or more embodiments of the invention, "first-area / second-area devices (202)" are any network devices configured to participate, at least partially, in both the first area (200) and the second area (200) of the network topology. As such, "first-area / second-area devices (202)" can be leaf devices or backbone devices within a backbone-leaf network topology.

[0041] An example of a device for a first area / second area is an area boundary device (e.g., area boundary device A (206), area boundary device N (208)). In one or more embodiments of the invention, an area boundary device (206, 208) is any network device in a network topology that is part of the first area (200) and has at least one connection to a device in the second area (210) that is not also in the first area (200) (e.g., a device located only in the second area). In one or more embodiments of the invention, an area boundary device (206, 208) comprises a first area connection state database (not shown) and a second area connection state database (not shown).In one or more embodiments of the invention, an area boundary device (206, 208) comprises functionality for sharing LSPs of the first area with first area devices and of LSPs of the second area with second area devices, and can populate its connection state databases using such LSPs received from other devices in the corresponding areas. In one or more embodiments of the invention, each area boundary device (206, 208) can directly or indirectly receive an area representation node identifier originating from an area conductor (204) inserted into the first area connection state database of the area boundary device.In addition, in one or more embodiments of the invention, each area boundary device (206, 208) maintains a proximity of the second area with the area conductor (204) either via a direct connection or via a tunnel implemented using any tunnel protocol.

[0042] Another example of a first-area / second-area device is an area conductor (204). In one or more embodiments of the invention, an area conductor (204) is a network device selected using any scheme for selecting a conductor within a network topology or a logical area of ​​a network topology. In one or more embodiments of the invention, an area conductor (204) has a second-area adjacency with any number of area boundary devices (206, 208) and is operationally connected to any number of second-area devices.In one or more embodiments of the invention, an area leader (204) is any network device chosen to represent the first area (200) by aggregating information from LSPs of the second area received by area boundary devices (206, 208), constructing an area representation node LSP using such information, and distributing the area representation node LSP among network devices in the second area, thereby injecting connection information relating to the area representation node into the connection state database of the second area from devices in the second area (210). The choice of the area leader (204) and the construction and distributing of the area representation node LSP are described further below. Fig. 3 discussed.

[0043] While Fig. 2. Although Figure 2 shows a configuration of components, other configurations may be used without deviating from the scope of the invention. For example, as discussed above, any number of devices in the first region and / or the second region, any number of first-region / second-region devices, any number of additional regions, and any number of external devices operationally connected to second-region devices may be present. Accordingly, embodiments disclosed herein should not be limited to the configuration of components shown in Figure 2. Fig. 2 are shown.

[0044] Fig. Figure 3 shows a flowchart describing a method for abstracting an area within a network that implements a link state routing protocol according to one or more embodiments of the invention.

[0045] In step 300, once one or more connection state databases, comprising connection information within one or more logical domains, have been established on each network device in a network topology (i.e., initial convergence has occurred), a domain leader is selected from among the network devices. In one or more embodiments of the invention, the domain leader selection can be performed using any scheme for selecting a single network device from among the network devices in a network topology or a logical domain of a network topology. Such schemes can include any type of leader priority specification, which can be any identifier capable of specifying a relative priority level from among a group of network devices and related to which device should be a domain leader network device.

[0046] For example, each network device can indicate a leader priority, and the network device with the highest numerical leader priority can be chosen as the area leader. In one or more embodiments of the invention, if more than one network device announces the same highest leader priority, a choice can be made between such devices using any other identifying information of the network devices. For example, from among the network devices that announce the same highest leader priority, the chosen leader can be the network device with the highest numerical system identifier, the Internet Protocol address, the router identification, etc.

[0047] In step 302, the area manager generates an area representation node identifier. In one or more embodiments of the invention, an area representation node identifier identifies an area representation node that is used as an abstraction of a first area of ​​a network topology. For example, an area representation node identifier can be a number represented by a series of binary (i.e., 1 or 0) bits that may be contained in a type-length-value (TLV) encoded section of an LSP of the first area.

[0048] In step 304, the area manager distributes the area representation node identifier to any number of area boundary devices. In one or more embodiments of the invention, the area representation node identifier, as part of a first area LSP, is distributed to first area devices adjacent to the area manager. Since the first area devices implement a link-state routing protocol such as IS-IS, they store the area representation node identifier in a first area link-state database and redistribute the first area LSP containing the area representation node identifier to adjacent first area devices that are different from the one from which the LSP was received.Thus, each device in the first region eventually receives the LSP and stores the region representation node identifier in its corresponding first region connection state databases. Since the region boundary devices also participate in the first region, they necessarily also receive the LSP and store the region representation node identifier in a first region connection state database. In other words, the region boundary devices can receive the LSP containing the region representation node identifier either directly from the region leader, if they happen to be first regions adjacent to the region leader, or indirectly as members of the first region connection state routing protocol scheme. Therefore, the region leader can be considered to indicate the existence of the region representation node to the region boundary devices via a first region LSP that includes the region representation node identifier.

[0049] In step 306, based on an identification of the existence of a region representation node identifier in a first region connection state database, each region constructs a second region LSP to be distributed to the region leader, to which the region boundary devices have either a direct or a tunneled connection, as discussed above. In one or more embodiments of the invention, the second region LSP comprises the region boundary device's boundaries to any second region devices and to the region leader, as well as the region representation node identifier previously received from the region leader (see step 304 above). Each region boundary device then transmits its corresponding second region LSP to the region leader.

[0050] In step 308, the area manager uses the LSPs received from the various area boundary devices of second areas to generate an area representation node LSP. In one or more embodiments of the invention, the area representation node LSP is generated by constructing an LSP that includes the area representation node identifier and all second area device boundaries of the various area boundary devices. Accordingly, the area representation node LSP can represent all boundaries of the first area to devices in the second area.

[0051] In step 310, the area representation node LSP generated in step 308 is split among second area devices. In one or more embodiments of the invention, the area representation node LSP is split using any splitting scheme of any connection state routing protocol. For example, the area representation node LSP can be flooded throughout the second area by successive splits to neighbors adjacent to the second area, starting with the area leader. Each second area device receiving the area representation node LSP can then store the information contained therein in corresponding second area connection state databases. The second area devices can then indicate a proximity to the area representation node instead of any specific one or more area boundary devices.Accordingly, any network traffic data unit received by a second area device destined for a device reachable through the network topology can be routed through the area representation node (i.e., the abstract first area). For example, a network traffic data unit can be received by a second area device and then routed to an area boundary device. The area boundary device can then use segment routing to route the network traffic data unit through the first area to another area boundary device, from which the network traffic data unit can further be routed to its final destination.

[0052] In step 312, it is determined whether the area leader has failed. In one or more embodiments of the invention, the area leader can be determined as failed using any scheme for determining that a network device has failed. For example, the area leader can be determined as failed by one or more other devices in a network topology that stop receiving expected communication (e.g., periodic packets) from the area leader. In one or more embodiments of the invention, an area leader must be present to create and partition the area representation node LSP. If the area leader has failed, the process continues to step 314. If the area leader has not failed, the process terminates. Although the basic process in Fig. Since the process is depicted in Figure 3 such that it terminates if the area manager has not failed, a person skilled in the art will recognize that the process can indeed continue as long as the network topology exists and is configured to implement one or more embodiments of the invention. Accordingly, the area manager can continue to periodically receive second-area LSPs from the area boundary devices, generate the area representation node LSP, and distribute the area representation node LSP to the second-area devices, unless at a certain point it is determined that the area manager has failed.

[0053] If, in step 312, it is determined that the area leader has failed, a new area leader is selected in step 314. In one or more embodiments of the invention, the selection of the new area leader may be substantially similar to the selection of the area leader discussed above in the description of step 300. In one or more embodiments of the invention, once the new area leader has been selected, the process returns to step 308, and the area representation node LSP is generated and distributed to second area devices. Although not in Fig. As shown in Figure 3, the process can alternatively return to step 302 to begin the process of creating an area representation node LSP if step 312 determines that the area manager has failed.

[0054] Fig. Figure 4 shows an example according to one or more embodiments of the invention. The following example serves only for explanatory purposes and is not intended to limit the scope of the invention.

[0055] With reference to Fig. Section 4 considers a scenario in which a backbone-leaf network topology is configured to implement an IS-IS link-state routing protocol. As such, backbone devices are contained in an L1 area, leaf devices are contained in an L2 area, and some devices participate in both L1 and L2 and are referred to as area-boundary devices.

[0056] In such a scenario, the network topology includes at least one L1 device (400), a selected area leader (404), at least two area boundary devices (402, 406), and at least two L2 devices (408). The L1 device (400) is directly or indirectly connected to and communicates with other L1 devices (not shown) and with devices that participate in both L1 and L2. As such, the L1 device (400) is operationally connected to at least the area boundary devices (402, 406) and a previously selected area leader (404).

[0057] As in Fig. As shown in Figure 4, the connection between the area conductor (404) and each of the area edge devices (402, 406) represents both an operational (i.e., direct or indirect) L1 connection and either a direct or a tunneled L2 adjacent connection. Area edge device A (402) is L2 adjacent and connected to L2 device A (408). Area edge device B (406) is L2 adjacent and connected to L2 device B (410). L2 device A (408) is operationally connected to external device A (412). L2 device B (410) is operationally connected to external device B (414).

[0058] In this scenario, the area manager (404), as the elected area manager, has created an area representation node identifier, TLV-encoded the area representation node identifier, included the TLV encoding in an L1-LSP, and distributed the L1-LSP to adjacent L1 devices, so that the L1-LSP is eventually flooded throughout the entire L1 area, including the area boundary devices (402, 406), which store the area representation node identifier in their respective L1 connection state databases.

[0059] The area boundary devices (402, 406) identify the area representation node identifier in their L1 connection state database and respond by generating an L2 LSP. The L2 LSP for area boundary device A (402) includes the area representation node identifier and at least the proximity information indicating the proximity to the area leader (404) and the L2 device A (408). The L2 LSP for area boundary device B (404) includes the area representation node identifier and at least the proximity information indicating the proximity to the area leader (404) and the L2 device B (410).

[0060] Using the two L2 LSPs received from the area boundary devices (402, 406), the area leader (404) then creates an area representation node LSP. The area representation node LSP includes the area representation node identifier and information indicating a proximity to both L2 device A (408) and L2 device B (410). The area representation node LSP is then distributed as a single L2 LSP to all L2 devices (e.g., 408, 410), which include the information in their respective L2 connection state databases instead of information that pertains to the entire spine-leaf network topology. The L2 devices can then indicate their proximity to the area representation node for the purpose of routing network traffic data units through the spine-leaf network topology.

[0061] In this scenario, a network traffic data unit is received at L2 device A (408) from external device A (412). The network traffic data unit is destined for external device B (414) as its final destination. For example, the network traffic data unit could be a Hypertext Transfer Protocol (HTTP) request from a mobile device (i.e., external device A (412)) destined for a web server (i.e., external device B (414)) and encapsulated in an IP packet within a MAC frame.

[0062] The L2 device A (408) receives the network traffic data unit based on its proximity to the area representation node, which in turn is adjacent to the L2 device B (410). In other words, the area representation node (i.e., the abstracted L1 area) is a necessary intermediate step along the path to the external device B (414) (i.e., the destination of the network traffic data unit).

[0063] To get the network traffic data unit from L2 device A (408) to L2 device B (410), L2 device A (408) first transfers the network traffic data unit to area edge device A (402). Area edge device A (402) then adds a label stack to the network traffic data unit, each label in the stack being designed to direct the network traffic data unit along a predetermined path to L2 device B (410). In the Fig.In the simple example shown in Figure 4, the label stack comprises two labels for sending the network traffic data unit along a path that includes the L1 device (400) and the area edge device B (406), each of which picks up its corresponding label before transmitting the network traffic data unit along the path to the L2 device B. Thus, the L2 device B receives the network traffic data unit without the label stack and routes it appropriately to the external device B.

[0064] Embodiments of the invention described above allow a first area to be abstracted when it is contained in a connection state database of a second area. This abstraction can significantly reduce the memory load within the connection state database of the second area. This reduction is achieved because not all connection state information for the topology of the first area needs to be included in the connection state database of the second area in order to route the network traffic of the second area through the first area.

[0065] While the invention has been described in relation to a limited number of embodiments, the person skilled in the art, having the benefit of this disclosure, will recognize that other embodiments can be conceived which do not deviate from the scope of the invention disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 7,177,951 B1

[0003] Cited non-patent literature

[0000] Dynamic Flooding on Dense Graphs; draft-li-dynamic-flooding-04.txt“ von T. Li, Arista Networks, Internet Engineering Task Force (IETF), Nr. 4, 27. März 2018, S. 1-17

[0002]

Claims

[1] Routing protocol scope abstraction system, the system comprising: a network value comprising network devices, wherein the network comprises a first network area comprising first network devices and a second network area comprising second network devices; a selected area manager trained to distribute an area representation node link state packet (LSP) to inject link information regarding an area representation node representing the first network area to an area link state database of the second network devices in the second network area. [2] System according to claim 1, wherein the domain representation node LSP comprises a domain representation node identifier which serves as an abstraction of the first network domain. [3] System according to claim 2, wherein the area representation node LSP comprises neighborhood data associated with the first network area. [4] System according to claim 3, wherein each second area network device stores the area representation node identifier in association with neighborhood data associated with the first network area in an associated area link state database, such that each second device announces a neighborhood to the area representation node. [5] System according to claim 4, wherein every second area network device redirects traffic destined for network devices on the network through the area representation node, based on the area representation node identifier and neighborhood data in the associated area link state database. [6] System according to claim 5, wherein the neighborhood data are determined by second area LSPs received by an area boundary device having at least one connection to a first network device and at least one connection to a second network device. [7] System according to claim 1, wherein the first network devices provide a network connection between second network devices in the second network area. [8] System according to claim 7, wherein the first network area or the second network area are layers of the network that implement a link state redirection protocol. [9] Network device, comprising: a link state database; a network interface that provides a connection to a network comprising network devices, comprising a first network area comprising first network devices and a second network area comprising second network devices, wherein the network device is configured to: Receiving a domain representation node link state packet (LSP) from a domain leader; and based on the reception of the area representation LSP, storing link information regarding an area representation node representing the first network area to the link state database of the network device. [10] Network device according to claim 9, wherein the domain representation node LSP comprises a domain representation node identifier which serves as an abstraction of the first network domain. [11] Network device according to claim 10, wherein the area representation node LSP comprises neighborhood data associated with the first network area. [12] Network device according to claim 11, wherein the area network device stores the area representation node identifier in association with neighborhood data associated with the first network area in the link state database, such that every second device announces a neighborhood to the area representation node. [13] Network device according to claim 12, wherein the area network device redirects traffic destined for network devices on the network through the area representation node, based on the area representation node identifier and neighborhood data in the associated area link state database. [14] Network device according to claim 9, wherein the first network devices provide a network connection between second network devices in the second network area. [15] Network device according to claim 14, wherein the first network area or the second network area are layers of the network that implement a link state redirection protocol.

Images