Device and method for nested multi-topology routing
The introduction of a control and agent entity for dynamic topology management in MTR systems addresses the challenge of adapting to traffic variations, enhancing network reliability and efficiency by enabling the activation and deactivation of nested topologies based on link utilization, thus optimizing traffic routing and energy use.
Patent Information
- Application Number
- PCT/EP2023/064838
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-06-02
- Publication Date
- 2024-12-05
AI Technical Summary
Existing Multi-Topology Routing (MTR) systems in IP networks face challenges in dynamically managing topologies to adapt to traffic variations, leading to potential disruptions, increased latency, and packet losses due to the need for frequent updates and energy inefficiency.
A control entity and agent entity are introduced to dynamically activate or deactivate pre-calculated nested topologies based on link utilization thresholds, allowing for centralized or distributed management to optimize traffic routing and energy efficiency.
This approach enables efficient traffic management, reduces packet loss and network disruptions, and improves network performance and reliability by allowing topologies to be dynamically adjusted according to traffic demands, thereby optimizing resource utilization and energy consumption.
Smart Images

Figure EP2023064838_05122024_PF_FP_ABST
Abstract
Description
[0001]DEVICE AND METHOD FOR NESTED MULTI-TOPOLOGY ROUTING TECHNICAL FIELD The present disclosure relates to Internet Protocol (IP) networks, and particularly to Multi- Topology Routing (MTR) in IP networks. In order to improve the use of network resources and improve the reliability of the network, the disclosure proposes devices and methods for managing MTR in a network. The disclosure is specifically concerned with the dynamic activation / deactivation of topologies to follow traffic variations. BACKGROUND In IP networks, the routing of data packets is most of the time decided using an Interior Gateway Protocol (IGP) which is a type of protocol used for exchanging routing information between routers. The most popular IGP protocol is a link-state routing protocol called Open Shortest Path First (OSPF). In OPSF, routers periodically share with other routers Link States Advertisements (LSA) messages that contain positive link metrics for their adjacent links. Upon reception of LSA messages, routers store them in a Link State Data Base (LSDB) and process this information using a shortest path algorithm to maintain a Shortest Path Tree (SPT) towards all the other nodes. The outcome of this computation is the determination of a next hop for all possible destinations. This information is stored in the data plane of devices in a Forwarding Information Base (FIB) and used to forward packets. In order to increase routing options, to support different application needs or to better load balance traffic in the network, an extension of IGP called Multi-Topology Routing (MTR) has been proposed. In this case, multiple IGP instances, also called topologies, are running in parallel, each of them working with a different set of link metrics and maintaining a Shortest- Path Tree (SPT). Each topology can relate to different routing criteria (e.g., cost, hop, delay, etc.), and different nodes on the same topology may have different SPTs (when act as root nodes). As routers execute multiple IGP instances, each instance has its own IP prefixes and advertises its own link states. For many network operators, MTR is a key technology to improve the use of network resources and improve reliability. With multi-topology routing, different logical topologies can be defined and used simultaneously in the same physical network to provide specific paths for different types of services. It can also be used to isolate certain types of services from each other. Sometimes, it may be necessary to disable or enable certain topologies to minimize energy consumption, in response to network congestion or failures. However, disabling a topology in a telecommunication network can potentially cause disruptions or impact network performance. Indeed, when a topology is disabled, it can affect the routing paths of some flows, leading to increased latency or even packet losses. Therefore, a system that allows dynamic activates and deactivates MTR topologies to frequently update routing over time, is desired. SUMMARY In view of the above-mentioned disadvantages, embodiments of the present disclosure aim to enable dynamic (de)activation of network topologies to follow traffic variations. In particular, an objective is to deploy several network topologies which can be dynamically (de)activated to follow traffic variations to fit with a redundant sequence of traffic matrices. One aim is to protect traffic on the network to avoid packet loss or loops in the network. Another aim is to enable energy-efficient routing with centralized and distributed routing solutions. The objective is achieved by embodiments as provided in the enclosed independent claims. Advantageous implementations of the embodiments are further defined in the dependent claims. A first aspect of the disclosure provides a control entity for managing MTR in a network, wherein the network comprises a plurality of routers, and the control entity is configured to generate a set of topologies for the network, wherein the set of topologies comprises N topologies, wherein an (n+1)th topology of the set of topologies comprises all links of an nth topology of the set of topologies, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; generate a policy for each topology of the set of topologies, wherein the policy is for triggering an activation and / or a deactivation of the topology; and configure each of the plurality of routers with the set of topologies. This disclosure allows deploying several topologies which can be dynamically (de)activated to follow traffic variations to fit with a redundant sequence of traffic matrices. A main idea of this disclosure is to rely on pre-calculated nested topologies that can be activated and deactivated in a given order. These topologies are able to protect traffic on the network to avoid packet loss or loops in the network. Notably, the ability to disable topologies in a nested MTR environment can help network operators manage their traffic more efficiently and effectively, thus improving network performance and reliability. In an implementation form of the first aspect, the policy comprises at least one of a first threshold and a second threshold, wherein the first threshold is for triggering an activation of the topology, and the second threshold is for triggering a deactivation of the topology. The policy comprises rules that are used for deciding whether and / or when to (de)activate a topology. Optionally. the rules can be simple thresholds, e.g., thresholds on the Link Utilization (LU) of specific links or on the Maximum LU. Possibly, the same threshold may be used for triggering the activation and the deactivation of the topology. Alternatively, the thresholds for triggering the activation and the deactivation of the topology may have different values. In an implementation form of the first aspect, the control entity is further configured to determine whether to activate or deactivate the (n+1)th topology of the set of topologies in the network based on the policy for that topology and an indicator of utilization of links in the network, and provide an activation indication or a deactivation indication to each of the plurality of routers when it is determined to activate or deactivate the (n+1)th topology in the network. For topologies (de)activations, two types of implementations are proposed in this disclosure: centralized and distributed. In the centralized solution, the control entity may have a global view of the network on top of the Link statistics that are periodically collected. Thus, the control entity may take a global decision on whether to activate or deactivate the topologies, and inform the routers once a decision is taken. In an implementation form of the first aspect, the control entity is further configured to determine to activate the (n+1)th topology in the network when the utilization of the links in the network exceeds the first threshold of the (n+1)th topology, and / or determine to deactivate the (n+1)th topology in the network when the utilization of the links in the network is below the second threshold of the (n+1)th topology. Optionally, the thresholds to (de)activate topologies can be on maximum link utilization, the sum of threshold link utilization, or link utilization of some links. In an implementation form of the first aspect, the activation indication indicates the router to activate the (n+1)th topology in the router, and / or the deactivation indication indicates the router to deactivate the (n+1)th topology in the router. Possibly, the (de)activation indication may comprise a Topology ID to inform the router(s) which topology is to be (de)activated. In an implementation form of the first aspect, the indicator of utilization of the links in the network indicates information of one or more of: a maximum link utilization, a sum of threshold link utilization, and link utilization of one or more predefined links. The control entity may obtain information about the link utilizations in the network. This may be sent by all nodes (routers) of the network to a network manager where the control entity is located. In an implementation form of the first aspect, the control entity is further configured to provide the policy of each topology of the set of topologies to each of the plurality of routers, for the router to decide whether to activate or deactivate a topology of the set of the topologies in the network. In the distributed topologies (de)activations, the control entity can configure rules (policies) for the routers so that the routers decide themselves when and / or whether to (de)activate a topology. In an implementation form of the first aspect, the control entity is further configured to generate the set of topologies based on traffic predictions of the network. Notably, this disclosure allows deploying several topologies which can be dynamically (de)activated to follow traffic variations, such as to fit with a redundant sequence of traffic matrices. In an implementation form of the first aspect, the control entity is further configured to generate the set of topologies further based on one or more of: an energy model of the network and a capacity constraint of the network, a quality of service requirement of the network. For example, an energy model can be given as an input to design energy-efficient nested topologies. When generating topologies, additional constraints or objectives can be considered. For instance, to ensure end-to-end QoS requirements in terms of latency, packet loss, the nested topologies may be generated based on a capacity constraint of the network, and / or a quality of service requirement of the network. In an implementation form of the first aspect, the policy of each topology further comprises a hysteresis parameter, wherein the hysteresis parameter indicates to delay the activation and / or the deactivation of the topology after the activation and / or the deactivation of the topology is triggered. Optionally, hysteresis can be added to avoid changing topologies too frequently. In an implementation form of the first aspect, the first threshold of the (n+1)th topology is higher than the first threshold of the nth topology, and / or the second threshold of the (n+1)th topology is higher than the second threshold of the nth topology. For example, it is may be configured that Topology 1 is activated when the MLU is below 30%, Topology 2 is activated when the MLU is between 30% and 50%, and Topology 3 is activated when the MLU is between 50% and 75%. In an implementation form of the first aspect, the first threshold of the (n+1)th topology is not below the second threshold of the (n+1)th topology. Possibly, the threshold for activating a topology is slightly higher than the threshold for deactivating topology. For instance, a value of the second threshold of a topology may be 90% of a value of the first threshold of the topology. However, it is also possible that the thresholds for activating and deactivating a topology have the same value. A second aspect of the disclosure provides an agent entity for assisting MTR in a network, wherein the agent entity is located at a router of the network, and the network comprises a plurality of routers, wherein the agent entity is configured to: obtain a set of topologies from a control entity, wherein the set of topologies comprises N topologies, wherein an (n+1)th topology of the set of topologies comprises all links of an nth topology of the set of topologies, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; and activate or deactivate a topology of the set of topologies in the router based on an activation indication or a deactivation indication from the control entity, or based on a policy for each topology of the set of topologies, wherein the policy indicates when to trigger an activation and / or a deactivation of the topology. An embodiment of this disclosure thus proposes an agent entity inside a router, which is in charge of the (de)activation of topologies and the graceful load balancing of traffic on top of them. The agent entity can either receive commands from the control entity about the (de)activation of topologies or decide locally based on rules (policies), which is also given by the control entity. In an implementation form of the second aspect, the agent entity is further configured to receive the activation indication or the deactivation indication from the control entity. In an implementation form of the second aspect, the agent entity is further configured to receive the policy for each topology of the set of topologies from the control entity. In an implementation form of the second aspect, the policy comprises at least one of a first threshold and a second threshold, wherein the first threshold indicates when to trigger an activation of the topology, and the second threshold indicates when to trigger a deactivation of the topology. In an implementation form of the second aspect, the agent entity is further configured to determine whether to activate or deactivate the (n+1)th topology of the set of topologies in the network based on the policy of that topology and an indicator of utilization of links in the network. The agent entity may (pre)obtain information about the link utilizations in the network. This information may be obtained through LSA exchanges between nodes (routers) of the network. In an implementation form of the second aspect, the agent entity is further configured to determine to activate the (n+1)th topology in the router when the utilization of the links in the network exceeds the first threshold of the (n+1)th topology, and / or determine to deactivate the (n+1)th topology in the router when the utilization of the links in the network is below the second threshold of the (n+1)th topology. In the distributed topologies (de)activations, the agent entity may receive periodically through LSAs information about link statistics in order to decide if a topology needs to be activated or deactivated, for instance, by checking the (de)activation thresholds of the topologies. In an implementation form of the second aspect, the agent entity is a first agent entity located at a first router, wherein the agent entity is further configured to send an activation notification or a deactivation notification to a second agent entity located at a second router of the plurality of routers in the network, wherein the activation notification or the deactivation notification indicates to the second agent entity that the first agent entity determined to activate or deactivate the (n+1)th topology in the first router. Once a (de)activation decision is taken by an agent entity of a router, this agent entity informs all other routers about the (de)activation decision. In an implementation form of the second aspect, the agent entity is further configured to receive an activation response or a deactivation response from the second agent entity, wherein the activation response or the deactivation response indicates to the first agent entity that the second agent entity determined to activate or deactivate the (n+1)th topology in the second router; and activate or deactivate the (n+1)th topology of the set of topologies in the first router. Before deactivating a topology, the agent entity needs to check whether this topology is not needed by any other routers. If it is not needed by all routers, every router deactivates this topology. When a message is received to activate a new topology (e.g., an activation indication from the control entity, or an activation response from another agent entity), the agent entity may wait for some time before using it so that all routers have activated it. In an implementation form of the second aspect, the policy of each topology further comprises a hysteresis parameter, wherein the hysteresis parameter indicates to delay the activation and / or the deactivation of the topology after the activation and / or the deactivation of the topology is triggered. Optionally, hysteresis can be added to avoid changing topologies too frequently. A third aspect of the disclosure provides a method for a control entity managing MTR in a network, wherein the method comprises: generating a set of topologies for the network, wherein the set of topologies comprises N topologies, wherein an (n+1)th topology of the set of topologies comprises all links of an nth topology of the set of topologies, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; generating a policy for each topology of the set of topologies, wherein the policy is for triggering an activation and / or a deactivation of the topology; and configuring each of the plurality of routers with the set of topologies. Implementation forms of the method of the third aspect may correspond to the implementation forms of the control entity of the first aspect described above. The method of the third aspect and its implementation forms achieve the same advantages and effects as described above for the control entity of the first aspect and its implementation forms. A fourth aspect of the disclosure provides a method for an agent entity assisting MTR in a network, wherein the agent entity is located at a router of the network, and the network comprises a plurality of routers, wherein the method comprises: obtaining a set of topologies from a control entity, wherein the set of topologies comprises N topologies, wherein an (n+1)th topology of the set of topologies comprises all links of an nth topology of the set of topologies, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; and activating or deactivating a topology of the set of topologies in the router based on an activation indication or a deactivation indication from the control entity, or based on a policy for each topology of the set of topologies, wherein the policy indicates when to trigger an activation and / or a deactivation of the topology. Implementation forms of the method of the fourth aspect may correspond to the implementation forms of the agent entity of the second aspect described above. The method of the fourth aspect and its implementation forms achieve the same advantages and effects as described above for the agent entity of the second aspect and its implementation forms. A fifth aspect of the disclosure provides a computer program product comprising a program code for carrying out, when implemented on a processor, the method according to the third aspect and any implementation forms of the third aspect, or the fourth aspect and any implementation forms of the fourth aspect. It has to be noted that all devices, elements, units and means described in the present application could be implemented in software or hardware elements or any kind of combination thereof. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements, or any kind of combination thereof. BRIEF DESCRIPTION OF DRAWINGS The above described aspects and implementation forms of the present disclosure will be explained in the following description of specific embodiments in relation to the enclosed drawings, in which: FIG.1 shows an example where different shortest path trees are calculated for different topologies; FIG.2 shows an example of a topology and shortest path trees with different root nodes; FIG.3 shows a control entity according to an embodiment of the disclosure; FIG.4 shows three nested topologies according to an embodiment of the disclosure; FIG.5 shows an agent entity according to an embodiment of the disclosure; FIG.6 shows a system architecture according to an embodiment of the disclosure; FIG.7 shows a topology of routers with link weights according to an embodiment of the disclosure; FIG.8 shows a procedure for nested MTR activation and deactivation according to an embodiment of the disclosure; FIG.9 shows a nested topology design according to an embodiment of the disclosure; FIG.10 shows a traffic scenario with variations according to an embodiment of the disclosure; FIG.11 shows an example with two topologies according to an embodiment of the disclosure; FIG.12 shows a method according to an embodiment of the disclosure; and FIG.13 shows a method according to an embodiment of the disclosure. DETAILED DESCRIPTION OF EMBODIMENTS Illustrative embodiments of a control entity, agent entity, and corresponding methods for managing MTR in a communication system are described with reference to the figures. Although this description provides a detailed example of possible implementations, it should be noted that the details are intended to be exemplary and in no way limit the scope of the application. Moreover, an embodiment / example may refer to other embodiments / examples. For example, any description including but not limited to terminology, element, process, explanation and / or technical advantage mentioned in one embodiment / example is applicative to the other embodiments / examples. In MTR, each topology is defined by a set of link weights and advertised through LSAs. A topology is used to define a routing between all nodes / routers which are involved by following the shortest path over the associated weights. MTR is compatible with multiple routing protocols like OSPF, Intermediate System to Intermediate Systems (ISIS), or FlexAlgo. FlexAlgo is an extension of MTR and adds the possibility to exploit a topology (i.e., a metric type) with different constraints (e.g., inclusions or exclusions of nodes / links) and a specific shortest path algorithm. Segment Routing (SR) can operate on top of IGP / MTR / FlexAlgo and it provides the possibility to concatenate shortest paths by using inflection points encoded as a label stack of IP addresses in packet headers for Segment Routing IPv6 (SRv6) protocol. MTR and FlexAlgo provide a good lightweight Traffic Engineering (TE) solution compared to Segment Routing (SR-TE) where a controller has to be deployed. For many network operators, MTR or FlexAlgo is a key technology to improve the use of network resources and improve reliability. It is commonly used as a lightweight traffic engineering solution in large-scale telecommunication networks where traffic types, such as voice, video, and data, have different requirements in terms of network bandwidth, delay, packet loss, etc. By creating multiple topologies, network operators can prioritize certain types of traffic and isolate them from each other, ensuring that network capacity is efficiently utilized and providing a better quality of service for end-users. Flows are generally mapped at ingress nodes to a specific topology depending on their requirements. FIG.1 illustrates three different topologies. From a root node, a different shortest path tree is calculated in each topology. These trees provide different routing paths from the root node to node c5. FIG.2 shows a topology and shortest path trees with different root nodes. As illustrated by FIG. 2, an MTR topology is defined by a set of weights on links (numbers in the figure) and so-called affinity rules to exclude some links (with “x” in the figure). Link weights represent the relative expense or effort required to transmit data across each link in the topology. The weights can be calculated based on various factors, such as link capacity, delay, jitter, and other quality of service (QoS) parameters, or specifically defined to obtain specific paths. The weights are used in each topology to compute the optimal path for flows between network nodes. Each node maintains a shortest path tree towards all the other nodes involved, considering potential link exclusions. Notably, routing needs to be continuously adjusted for various reasons in telecommunication networks to minimize congestion, satisfy end-to-end QoS requirements, or minimize energy consumption. For instance, to minimize energy consumption, traffic can be steered over time to control the utilization of links so that energy savings mechanisms deliver maximum efficiency. When minimizing the congestion, routing paths can be adjusted over time to minimize the Maximum Link Utilization (MLU). Multiple mechanisms exist for the dynamic adjustment of routing paths. Segment Routing is one of the most flexible mechanisms. Using labels, a source can design a new path by adding a set of inflection points. This technology generally requires the deployment of a centralized controller, which is not desirable in some cases for scalability, fault tolerance, or deployment reasons. When distributed IGP is used, one solution is to update link weights according to traffic variations. However, when the IGP topology is updated, routing loops can appear and traffic can be dropped. In practice, it is a (very) slow and risky operation. Network operators normally plan IGP weights for the long term and update them in a very careful and seldom manner. For distributed IGP, deploying and using different topologies (e.g., different sets of link weights for instance) to follow traffic variations (low, intermediate, or high traffic for instance) is not feasible in practice. Energy-aware routing and traffic engineering solutions have emerged as a natural complement to power management mechanisms. Their goal is to steer traffic so that energy gains can be maximized. In existing solutions, some mathematical models are proposed to decide which network element (ports, routers, links, or line cards) to switch off such that the traffic demand is satisfied and the total energy consumption is minimized. However, these solutions are either fully centralized or only manage a single IGP topology (limited flexibility as it needs to be slowly and carefully updated). This disclosure thus provides a system that is able to dynamically activate and deactivate pre- installed MTR topologies so as to frequently update routing over time. It provides the right balance between routing control and fully distributed operations. FIG.3 shows a control entity 300 adapted for managing MTR in a network 400, according to an embodiment of the disclosure. The control entity 300 may comprise processing circuitry (not shown) configured to perform, conduct or initiate the various operations of the control entity 300 described herein. The processing circuitry may comprise hardware and software. The hardware may comprise analog circuitry or digital circuitry, or both analog and digital circuitry. The digital circuitry may comprise components such as application-specific integrated circuits (ASICs), field-programmable arrays (FPGAs), digital signal processors (DSPs), or multi- purpose processors. The control entity 300 may further comprise memory circuitry, which stores one or more instruction(s) that can be executed by the processor or by the processing circuitry, in particular under the control of the software. For instance, the memory circuitry may comprise a non-transitory storage medium storing executable software code which, when executed by the processor or the processing circuitry, causes the various operations of the control entity 300 to be performed. In one embodiment, the processing circuitry comprises one or more processors and a non-transitory memory connected to the one or more processors. The non-transitory memory may carry executable program code which, when executed by the one or more processors, causes the control entity 300 to perform, conduct or initiate the operations or methods described herein. In particular, the network 400 comprises a plurality of routers 401. The control entity 300 is configured to generate a set of topologies 301 for the network 400. The set of topologies 301 comprises N topologies, wherein an (n+1)th topology of the set of topologies 301 comprises all links of an nth topology of the set of topologies 301, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N. The control entity 300 is further configured to generate a policy 302 for each topology of the set of topologies 301. Notably, the policy 302 is for triggering an activation and / or a deactivation of the topology. Then, the control entity 300 is configured to configure each of the plurality of routers 401 with the set of topologies 301. A main idea of this disclosure is to rely on pre-calculated nested topologies that can be activated and deactivated in a given order. Topologies are designed as Russian dolls: when topology N is active, the topologies of index <N are also active. This disclosure allows deploying several topologies which can be dynamically (de)activated to follow traffic variations to fit with a redundant sequence of traffic matrices. These topologies are able to protect traffic on the network to avoid packet loss or loops in the network. Conventionally, the IGP topology needs to be updated too frequently, which is a risky and slow operation where weights have to be updated in a very incremental manner and performance degradations can happen. This disclosure avoids updated link weights of a single IGP topology. FIG.4 shows an example of 3 nested topologies according to an embodiment of this disclosure, where Topology 1 is included in Topology 2 and Topology 2 is included in Topology 3. It must be noted that when topologies are nested, it becomes easier to enable and disable specific topologies without causing significant disruptions to network traffic. In particular, when topology N is deactivated, all the traffic routed over topologies <N does not change. Furthermore, traffic routed over topology N represents only a small fraction of the total traffic and can be gracefully re-routed over topologies with index <N before topology N is deactivated. Overall, the ability to disable topologies in a nested MTR environment can help network operators manage their traffic more efficiently and effectively, thus improving network performance and reliability. A goal of such a nested topology design is to activate a topology to absorb the traffic increase and to deactivate it when traffic decreases. Optionally, according to an embodiment of the disclosure, the policy 302 comprises at least one of a first threshold and a second threshold, wherein the first threshold is for triggering an activation of the topology, and the second threshold is for triggering a deactivation of the topology. For topologies (de)activations, two types of implementations are proposed in this disclosure: centralized and distributed. In the centralized solution, the control entity 300 has a global view of the network on top of the Link statistics that are periodically collected. It thus be able to take a global decision in order to activate or deactivate the topologies. Optionally, according to an embodiment of the disclosure, the control entity 300 may be configured to determine whether to activate or deactivate the (n+1)th topology of the set of topologies 301 in the network 400 based on the policy 302 for that topology and an indicator of utilization of links in the network 400. Once a decision is taken, it is transmitted to the routers. The control entity 300 may be further configured to provide an activation indication 303 or a deactivation indication 304 to each of the plurality of routers 401 when it is determined to activate or deactivate the (n+1)th topology in the network 400. According to an embodiment of the disclosure, the control entity 300 may be configured to determine to activate the (n+1)th topology in the network 400 when the utilization of the links in the network 400 exceeds the first threshold of the (n+1)th topology. Optionally, according to an embodiment of the disclosure, the control entity 300 may be configured to determine to deactivate the (n+1)th topology in the network 400 when the utilization of the links in the network 400 is below the second threshold of the (n+1)th topology. The control entity 300 may (pre)obtain information about the link utilizations in the network. This may be sent by all nodes of the network to a network manager where the control entity 300 is located. For example, the indicator of utilization of the links in the network 400 may indicate information of one or more of: a maximum link utilization, a sum of threshold link utilization, and link utilization of one or more predefined links. Optionally, the activation indication 303 may indicate the router 401 to activate the (n+1)th topology in the router 401. Similarly, the deactivation indication 304 may indicate the router 401 to deactivate the (n+1)th topology in the router 401. In the distributed solution, necessary information may be provided to the routers by the control entity 300, for each router to decide if a topology needs to be activated or deactivated. According to an embodiment of the disclosure, the control entity 300 may be configured to provide the policy 302 of each topology of the set of topologies 301 to each of the plurality of routers 401, for the router to decide whether to activate or deactivate a topology of the set of the topologies 301 in the network 400. According to an embodiment of the disclosure, the control entity 300 may be configured to generate the set of topologies 301 based on traffic predictions of the network 400. According to an embodiment of the disclosure, the control entity 300 may be configured to generate the set of topologies 301 further based on one or more of: an energy model of the network 400 and a capacity constraint of the network 400, a quality of service requirement of the network 400. According to an embodiment of the disclosure, the policy 302 of each topology further comprises a hysteresis parameter, wherein the hysteresis parameter indicates to delay the activation and / or the deactivation of the topology after the activation and / or the deactivation of the topology is triggered. Optionally, the first threshold of the (n+1)th topology is higher than the first threshold of the nth topology, and / or the second threshold of the (n+1)th topology is higher than the second threshold of the nth topology. Notably, the first threshold of the (n+1)th topology may not below the second threshold of the (n+1)th topology. FIG.5 shows an agent entity 500 adapted for assisting MTR in a network 400 according to an embodiment of the disclosure. The agent entity 500 may comprise processing circuitry (not shown) configured to perform, conduct or initiate the various operations of the agent entity 500 described herein. The processing circuitry may comprise hardware and software. The hardware may comprise analog circuitry or digital circuitry, or both analog and digital circuitry. The digital circuitry may comprise components such as application-specific integrated circuits (ASICs), field-programmable arrays (FPGAs), digital signal processors (DSPs), or multi- purpose processors. The agent entity 500 may further comprise memory circuitry, which stores one or more instruction(s) that can be executed by the processor or by the processing circuitry, in particular under control of the software. For instance, the memory circuitry may comprise a non-transitory storage medium storing executable software code which, when executed by the processor or the processing circuitry, causes the various operations of the agent entity 500 to be performed. In one embodiment, the processing circuitry comprises one or more processors and a non-transitory memory connected to the one or more processors. The non-transitory memory may carry executable program code which, when executed by the one or more processors, causes agent entity 500 to perform, conduct or initiate the operations or methods described herein. In particular, the network 400 may comprise a plurality of routers 401. The agent entity 500 is located at a router 401 of the network 400. The agent entity 500 is configured to obtain a set of topologies 301 from a control entity 300. In particular, the set of topologies 301 comprises N topologies, wherein an (n+1)th topology of the set of topologies 301 comprises all links of an nth topology of the set of topologies 301, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N. The agent entity 500 is further configured to activate or deactivate a topology of the set of topologies 301 in the router 401 based on an activation indication 303 or a deactivation indication 304 from the control entity 300. Alternatively, the agent entity 500 is configured to activate or deactivate a topology of the set of topologies 301 in the router 401 based on a policy 302 for each topology of the set of topologies 301, wherein the policy 302 indicates when to trigger an activation and / or a deactivation of the topology. As previously discussed, for topologies (de)activations, two types of implementations are proposed in this disclosure. One is a centralized solution, where the control entity 300 takes a global decision on whether to activate or deactivate a topology. Once a decision is taken, the control entity 300 notifies the router(s) 401 in the network 400 by providing an activation indication 303 or a deactivation indication 304. Optionally, the agent entity 500 may be further configured to receive the activation indication 303 or the deactivation indication 304 from the control entity 300. Then, the agent entity 500 activates or deactivates a topology accordingly. The other solution is a distributed solution, where the agent entity 500 makes the decision based on the policy 302 of each topology obtained from the control entity 300. In this case, the agent entity may be configured to receive the policy 302 for each topology of the set of topologies 301 from the control entity 300. Optionally, the policy 302 may comprise at least one of a first threshold and a second threshold, wherein the first threshold indicates when to trigger an activation of the topology, and the second threshold indicates when to trigger a deactivation of the topology. Optionally, according to an embodiment of the disclosure, the agent entity 500 may be further configured to determine whether to activate or deactivate the (n+1)th topology of the set of topologies 301 in the network 400 based on the policy 302 of that topology and an indicator of utilization of links in the network 400. The agent entity 500 may obtain information about the link utilizations in the network. This information may be obtained through LSA exchanges between nodes of the network. For example, the indicator of utilization of the links in the network 400 may indicate information on one or more of: a maximum link utilization, a sum of threshold link utilization, and link utilization of one or more predefined links. Optionally, according to an embodiment of the disclosure, the agent entity may be configured to determine to activate the (n+1)th topology in the router when the utilization of the links in the network 400 exceeds the first threshold of the (n+1)th topology. Optionally, the agent entity may be configured to determine to deactivate the (n+1)th topology in the router when the utilization of the links in the network 400 is below the second threshold of the (n+1)th topology. In the distributed case, once the agent entity 500 takes a decision to activate or deactivate a topology, it informs all other routers in the network. According to an embodiment of the disclosure, the agent entity 500 is a first agent entity 500 located at a first router 401. , The first agent entity 500 is configured to send an activation notification or a deactivation notification to a second agent entity 500’ located at a second router 401’ of the plurality of routers 401 in the network 400. In particular, the activation notification or the deactivation notification indicates to the second agent entity 500’ that the first agent entity 500 determined to activate or deactivate the (n+1)th topology in the first router 401. According to an embodiment of the disclosure, the first agent entity 500 may be further configured to receive an activation response or a deactivation response from the second agent entity 500’. The activation response or the deactivation response indicates to the first agent entity 500 that the second agent entity 500’ determined to activate or deactivate the (n+1)th topology in the second router 401’. After the activation response or the deactivation response is received at the first agent entity 500, the agent entity 500 is further configured to activate or deactivate the (n+1)th topology of the set of topologies 301 in the first router 401. Optionally, the policy 302 of each topology further comprises a hysteresis parameter, wherein the hysteresis parameter indicates to delay the activation and / or the deactivation of the topology after the activation and / or the deactivation of the topology is triggered. FIG. 6 depicts a system architecture according to an embodiment of the disclosure. This architecture comprises a control entity 300 and an agent entity 500 located in a router 401. In particular, the control entity 300 may be the control entity shown in FIG.3, and the agent entity 500 may be the agent entity shown in FIG.5. The control entity 300 may be located in a network manager. As shown in FIG.6, this disclosure relies on two specific modules as illustrated by FIG.6: 1) a control entity 300, which may also be named as a “Nested MTR Config” module (NMC), inside the Network Manager, and 2) an agent entity 500, namely, a “Nested MTR Agent” module (NMA) inside a router. The NMC module is in charge of pre-calculating the topologies and installing them in routers. Possibly, each topology can be an MTR topology with a set of link weights or a FlexAlgo (link weights, constraints on links exclusion, and the shortest path algorithm). Link weights can be configured using Command Line Interface (CLI) or the Netconf protocol. Flexalgo can be described using the FlexAlgo Definition (FAD) message. The NMC module can decide about the activation or deactivation of topologies and enforce its decisions in routers. Alternatively, it can configure rules (i.e., policies, such as the policy 302 as shown in FIG.3) so that routers (e.g., the router 401) decide when to activate or deactivate them. The NMA module is in charge of the activation or deactivation of topologies and the graceful load balancing of traffic on top of them. It can either receive commands from the NMC module about the activation or deactivation of topologies or decide locally based on rules, which may be given by the NMC module. Based on traffic information (e.g., traffic predictions), the NMC module can determine K nested topologies (e.g., the set of topologies 301) and their associated (de)activation policies (e.g., thresholds, for example, the MLU). K is a positive integer and can be given as input by a network operator and might be related to memory limitations in routers. An ordered set of K representative traffic matrices can be given in input with a specific traffic engineering objective, like the minimization of the MLU. In case that energy-consumption needs to be minimized, information about the energy consumption of devices may also be given to the NMC module. (De)activation rules (e.g., the policy 302 as shown in FIG.3) can be transmitted to the Nested MTR agents of all routers in the network, using the Nested MTR Configurations message, i.e., NM-C message. The rules can be simple thresholds on the Link Utilization (LU) of specific links or on the Maximum LU. Optionally, the NM-C message can comprise the following information: • Topology IDs • Activation thresholds For topologies (de)activations, two types of implementations are possible: centralized and distributed. In the centralized solution, the NMC module has a global view of the network on top of the Link statistics that are periodically collected. It takes a global decision in order to activate or deactivate the topologies. Once the decision is taken, they are transmitted to the NMA modules via the “Nested MTR Topology Activation” message (NM-TA-1). In the distributed case, NMA modules can receive also periodically through LSAs information about link statistics in order to decide if a topology needs to be activated or deactivated, by checking the activation thresholds of the topologies. Once a decision is taken, the NMA module informs all other routers using an Activation (NM-TA-2) message. Optionally, the NM-TA message can comprise the following information: • Topology ID • Activate / deactivate Possibly, a new flag in LSAs defined as a Type Length Value (TLV) field can be used by routers to distinguish pre-calculated nested topologies from others. This new TLV in LSAs that flags the nested topology tells the router that the shortest path trees can be precomputed without deploying forwarding rules (routes). FIG. 7 shows a topology of routers with link weights according to an embodiment of this disclosure. According to this example, traffic predictions and an energy model are given to the NMC module to determine K topologies and their associated activation thresholds. FIG. 7 shows the outcome of the computation. Topology 1 is composed of solid links and is activated when the MLU is below 30%. Topology 2 is composed of solid lines and dashed lines and is activated when the MLU is below 50%. Topology 3 is composed of solid lines, dashed lines, and dotted lines and is activated when the MLU is below 75%. For instance, when the MLU increases above 30%, an NM-TA is sent to activate Topology 2. When the traffic decreases and the MLU falls below 30%, a new NM-TA is sent. Notably, additional mechanisms can be added for stability, which will be explained in the following part of the application. FIG. 8 shows a procedure for nested MTR activation and deactivation according to an embodiment of this disclosure. The overall process comprises the following steps: Step 1: Reception of nested MTR configuration from the Controller • Based on traffic predictions, the NMC module (i.e., the control entity 300 as shown in FIG.3) in the controller (e.g., Network manager as shown in FIG.7) configures K topologies and determines (de)activation policies (e.g., thresholds (details are discussed in FIG.9) • The NMC module configures MTR topologies inside routers (e.g., using Netconf, CLI, etc.) • The NMC module provides NMA modules (i.e., the agent entity 500 as shown in FIG. 5) with the policies to (de)activate topologies (e.g., using NM-C msg.). This is only done for the distributed setting where routers (i.e., NMA modules) make decisions about (de)activations. Step 2: Decide whether to (de)activate topologies • In the distributed setting, routers can decide whether to (de)activate topologies based on (de)activation policies. • In the centralized setting, the controller decides whether to (de)activate topologies based on (de)activation policies, • In both cases, the decision is communicated to all routers (e.g., using NM-TA msgs). Step 3: Gracefully (de)activate topologies inside devices • When an NM-TA msg is received by the NMA module inside each device, the load balancing for ingress traffic is first notified to re-organize traffic on remaining topologies. For instance, make-before-break (MBB) or “flowlets” may be used to avoid packet losses. • Then, the status of the newly (de)activated topologies is updated. When a topology (IGP instance) is deactivated, routes associated with this topology are deleted and LSAs stop being advertised. When a new topology is activated, the routes are automatically added and a portion of the traffic is using them. Go back to Step 2 after Step 3 is done. Notably, for the distributed setting, a specific signal (i.e., NM-C msg) is needed to inform other routers which topology is activated. It depends on the activation threshold. The shortest path trees associated with topologies are precomputed and ready to use. FIG. 9 shows an example of nested topology design according to an embodiment of the disclosure. In particular, three different traffic matrices (scenarios) are given as input, ranked by the global traffic (e.g., diurnal or regional traffic) sizes. The NMC module designs three nested topologies (K=3) such that the MLU is minimized for every scenario. In this example, the MLU for each scenario are used as the (de)activation threshold for each computed topology. In this case, the following mathematical model is used for computing |^ | topologies. The topologies are pairwise included. It aims to route traffic of every matrix ^ on the associated topologies ^^taking into account the capacity constraints. It also computes the OSPF weights and the threshold MLUs. The model may include the following inputs, variables, and linear programs. Inputs: • ^: set of topology IDs • ^: set of traffic matrices • ^: set of traffic matrices, |^| = |^| • ^^⊂ ^^… ⊂ ^|^|= ^: set topologies for every traffic matrix • ^: set of destination nodes • ^: set of links in the network Variables: • ^^^^^: traffic of tunnel ^ of matrix ^, routed on link ^ ∈ ^ of topology ^ ∈ ^. • ^^^^= 1 if link ^ belongs to tree of destination ^ ∈ ^, in topology ^ ∈ ^. • ^^^^: OSPF weight on link(^^)∈ A for topology ^ ∈ ^. • ^^^= 1 if link ^ ∈ ^ is used in topology ^ ∈ ^.0 otherwise. • ^^= 1 if link ^ ∈ ^ is used.0 otherwise. • ^^^^: MLU threshold of matrix ^. Integer linear program: Routing objective: Flow conservation: ∈ ^, ∀^ ∈ ^, 0 ^^ℎ^^^^^^ Threshold computation : ^ ^^^^+ ^ ≤ ^^^^^^∀^ ∈ ^\{|^|} where ^ is a parameter providing enough difference between two thresholds. Tree definition: Nested topology constraints: ^ ^^^^≤ ^^^∀^ ∈ ^, ∀^ ∈ ^, ∀^ ∈ ^, ^ ^^^≤ ^^^^^∀^ ∈ ^, ∀^ ∈ ^, ∀^ ∈ ^, ^ variables definition: ^ ^^^≤ ^^∀^ ∈ ^, ∀^ ∈ ^, OSPF weights: ^ ^^^^− ^^^^+ ^^^^≥ 1 − ^^^^∀^ ∈ ^, ∀^ ∈ ^, ∀^ ∈ ^, where ℤ^presents the set of all positive integers. In a particular use case, the above model can be used for energy optimization (i.e., energy consumption minimization) as follows: ^ Get a set of traffic matrices ranked with an ascending order of the global traffic amount. ^ Define the cost vector “^^^^_^^^^_^^^^" as fixed positive activation link costs. These costs will avoid activating useless links. ^ Define the cost vector “ ^^^^^^_^^^^_^^^^ ” which corresponds to the energy consumption of the links. The energy consumption must be penalized with a strictly increasing function ^(^), where ^ ∈ ^ is the topology ID (for example ^(^) = ^^). Thus, the first topology may avoid energy-consuming links. ^ Define the “ ^^^_^^^^ ” that pushes the model to minimize the maximum link utilization for every topology that will be used later as a (de)activation threshold. Notably, the model will compute |^| nested topologies such that the first topology (the one with the least number of links) includes the low-energy consumption links. Once traffic increases, based on the MLU threshold of the second topology, the second topology will be activated leading to the turn-on of new links in the network. This process will be repeated if the traffic further increases. It may be understood that a goal of the above-discussed algorithm provides three topologies such that the associated support networks are embedded, ^^1 ⊆ ^^2 ⊆ ^^3, as shown in FIG. 9 (the support network of each topology is composed of the links represented in bold lines). Notably, an energy model can be given as an input to design energy-efficient nested topologies. For example, the nested topologies (e.g., the set of topologies 301 as shown in FIG.3) may be generated based on an energy model of the network. When generating topologies, additional constraints or objectives can be considered. For instance, to ensure end-to-end QoS requirements in terms of latency and packet loss, the nested topologies may be generated based on link capacity constraints, and / or a quality of service requirement of the network. According to an embodiment of the disclosure, a set of thresholds can be used for topologies activation and deactivation. FIG.10 shows a traffic scenario with variations. For instance, a Nested MTR Topology Activation message (NM-TA msg) can be sent by the controller (e.g., the control entity 300 as shown in FIG.3) or routers (e.g., the agent entity 500 located in the router as shown in FIG.5) in the following cases: • If the MLU increases above the threshold, then an NM-TA is sent and a new topology can be used. • If the MLU decreases below the threshold safeguard, then an NM-TA is sent and a topology is deactivated. Possibly, a threshold that is set for the activation of a topology and a threshold that is set for the deactivation of a topology may have different values. As shown in FIG.10, the threshold for activating topology 1 (or topology 2) is slightly higher than the threshold for deactivating topology 1 (or topology 2). For instance, this may have the effect that topologies will not be activated and deactivated frequently. However, it is also possible that the thresholds for activating and deactivating a topology have the same value. In this embodiment, the MLU threshold is only used as one possible example. The thresholds may be set based on other factors. Optionally, the threshold to activate topologies may be designed using MLU, a sum of threshold link utilization, or link utilization of some links. In addition, hysteresis can be added to avoid changing topologies too frequently. Possibly, additional topologies can be considered to ensure specific SLAs (at the cost of lower energy efficiency). This present disclosure may also apply for multi-domain IGP, in this case, topology management and (de)activation policies (e.g., thresholds) can remain inside each area. When an NM-TA msg is received to activate a new topology, the router may wait for some time before using it so that all routers have activated it. This time may be computed according to the size of the network and / or the time required to turn on devices. When an NM-TA msg is received to deactivate a topology, no new traffic is sent on this topology. Notably, when a topology is not needed anymore, every node (router) will deactivate it. First, nodes stop sending traffic on this topology. The topology can stay active for some time until flow finishes, alternatively, the ingress load balancer can proactively re-organize traffic. FIG.11 shows an example with two topologies (topo1, topo2) according to an embodiment of the disclosure. Each of the nodes in this example may be a router 401 as shown in FIG.3, FIG. 5, or FIG.6. The weights of a link of topo 1 and topo 2 are shown in FIG.11. For example, the weights for the link between Node A and Node B are shown as (∞, 8), the first number represents the weight in topo 1, and the second number represents the weight in topo 2. In this example, “∞” means that this link is switched off for topo 1. It is supposed that every router has, for every topology t, a Queue list, that contains the ID of routers that have required the topology t. When a router requires a topology, it sends a signal to all nodes, even if the topology is already activated. For instance, in Step 1, Node A realizes that it does not need topology 2. In Step 2, Node A informs all other nodes, e.g., using NM-TA msgs. In Step 3, each node checks whether all nodes that previously asked for topology 2 sent a signal to deactivate it. In Step 4, if topology 2 is not needed by any node, every router deactivates topology 2. FIG.12 shows a method 1200 according to an embodiment of the disclosure, particularly for managing MTR in a network 400. In particular, the network 400 comprises a plurality of routers 401. In a particular embodiment, the method 1100 is performed by a control entity 300 shown in FIG. 3, FIG. 5, or FIG. 6. The method 1200 comprises a step 1201 of generating a set of topologies 301 for the network 400. The set of topologies 301 comprises N topologies, wherein an (n+1)th topology of the set of topologies 301 comprises all links of an nth topology of the set of topologies 301, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N. The method 1200 further comprises a step 1202 of generating a policy 302 for each topology of the set of topologies 301, wherein the policy 302 is for triggering an activation and / or a deactivation of the topology. Further, the method 1200 comprises a step 1203 of configuring each of the plurality of routers 401 with the set of topologies 301. Possibly, the router 401 is the router shown in FIG.3, FIG.5, FIG.6, or FIG.7. FIG.13 shows a method 1300 according to an embodiment of the disclosure, particularly for assisting MTR in a network 400. In a particular embodiment, the method 1300 is performed by an agent entity 500 shown in FIG.5. In particular, the network comprises a plurality of routers 401, and the agent entity 500 is located at a router 401 of the network 400. The method 1300 comprises a step 1301 of obtaining a set of topologies 301 from a control entity 300. The set of topologies 301 comprises N topologies, wherein an (n+1)th topology of the set of topologies 301 comprises all links of an nth topology of the set of topologies 301, wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N. The method 1300 further comprises a step 1302 of activating or deactivating a topology of the set of topologies 301 in the router 410 based on an activation indication 303 or a deactivation indication 304 from the control entity 300, or based on a policy 302 for each topology of the set of topologies 301. The policy 302 indicates when to trigger an activation and / or a deactivation of the topology. Possibly, the control entity 300 is the control entity shown in FIG.3, FIG.5, or FIG.6, and the router 401 is the router shown in FIG.3, FIG.5, FIG.6 or FIG.7. In the disclosure, apparatus and methods for nested MTR in IP networks are proposed. Embodiments of the disclosure allows deploying several topologies which can be dynamically (de)activated to follow traffic variations to fit with a redundant sequence of traffic matrices. These topologies are able to protect traffic on the network to avoid packet loss or loops in the network. Conventionally, the IGP topology needs to be updated too frequently, which is a risky and slow operation where weights have to be updated in a very incremental manner and performance degradations can happen. This disclosure avoids updated link weights of a single IGP topology. The present disclosure has been described in conjunction with various embodiments as examples as well as implementations. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed embodiments of the disclosure, from the studies of the drawings, this disclosure and the independent claims. In the claims as well as in the description the word “comprising” does not exclude other elements or steps and the indefinite article “a” or “an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutual different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation. Furthermore, any method according to embodiments of the disclosure may be implemented in a computer program, having code means, which when run by processing means causes the processing means to execute the steps of the method. The computer program is included in a computer readable medium of a computer program product. The computer readable medium may comprise essentially any memory, such as a ROM (Read-Only Memory), a PROM (Programmable Read-Only Memory), an EPROM (Erasable PROM), a Flash memory, an EEPROM (Electrically Erasable PROM), or a hard disk drive. Moreover, it is realized by the skilled person that embodiments of the control entity 300, or the agent entity 500, comprises the necessary communication capabilities in the form of e.g., functions, means, units, elements, etc., for performing the solution. Examples of other such means, units, elements and functions are: processors, memory, buffers, control logic, encoders, decoders, rate matchers, de-rate matchers, mapping units, multipliers, decision units, selecting units, switches, interleavers, de-interleavers, modulators, demodulators, inputs, outputs, antennas, amplifiers, receiver units, transmitter units, DSPs, trellis-coded modulation (TCM) encoder, TCM decoder, power supply units, power feeders, communication interfaces, communication protocols, etc. which are suitably arranged together for performing the solution. Especially, the processor(s) of the control entity 300, or the agent entity 500 may comprise, e.g., one or more instances of a Central Processing Unit (CPU), a processing unit, a processing circuit, a processor, an Application Specific Integrated Circuit (ASIC), a microprocessor, or other processing logic that may interpret and execute instructions. The expression “processor” may thus represent a processing circuitry comprising a plurality of processing circuits, such as, e.g., any, some or all of the ones mentioned above. The processing circuitry may further perform data processing functions for inputting, outputting, and processing of data comprising data buffering and device control functions, such as call processing control, user interface control, or the like.
Claims
CLAIMS 1. A control entity (300) for managing multi-topology routing in a network (400), wherein the network (400) comprises a plurality of routers (401), and the control entity (300) is configured to: generate a set of topologies (301) for the network (400), wherein the set of topologies (301) comprises N topologies, wherein an (n+1)th topology of the set of topologies (301) comprises all links of an nth topology of the set of topologies (301), wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; generate a policy (302) for each topology of the set of topologies (301), wherein the policy (302) is for triggering an activation and / or a deactivation of the topology; and configure each of the plurality of routers (401) with the set of topologies (301).
2. The control entity (300) according to claim 1, wherein the policy (302) comprises at least one of a first threshold and a second threshold, wherein the first threshold is for triggering an activation of the topology, and the second threshold is for triggering a deactivation of the topology.
3. The control entity (300) according to claim 1 or 2, further configured to: determine whether to activate or deactivate the (n+1)th topology of the set of topologies (301) in the network (400) based on the policy (302) for that topology and an indicator of utilization of links in the network (400), and provide an activation indication (303) or a deactivation indication (304) to each of the plurality of routers (401) when it is determined to activate or deactivate the (n+1)th topology in the network (400).
4. The control entity (300) according to claims 2 and 3, further configured to: determine to activate the (n+1)th topology in the network (400) when the utilization of the links in the network (400) exceeds the first threshold of the (n+1)th topology, and / or determine to deactivate the (n+1)th topology in the network (400) when the utilization of the links in the network (400) is below the second threshold of the (n+1)th topology.
5. The control entity (300) according to claim 4, wherein the activation indication (303) indicates the router (401) to activate the (n+1)th topology in the router (401), and / or wherein the deactivation indication (304) indicates the router (401) to deactivate the (n+1)th topology in the router (401).
6. The control entity (300) according to one of the claims 3 to 5, wherein the indicator of utilization of the links in the network (400) indicates information of one or more of: a maximum link utilization, a sum of threshold link utilization, and link utilization of one or more predefined links.
7. The control entity (300) according to claim 1 or 2, further configured to: provide the policy (302) of each topology of the set of topologies (301) to each of the plurality of routers (401), for the router to decide whether to activate or deactivate a topology of the set of the topologies (301) in the network (400).
8. The control entity (300) according to one of the claims 1 to 7, further configured to: generate the set of topologies (301) based on traffic predictions of the network (400).
9. The control entity (300) according to claim 8, further configured to: generate the set of topologies (301) further based on one or more of: an energy model of the network (400), a capacity constraint of the network (400), a quality of service requirement of the network (400).
10. The control entity (300) according to one of the claims 1 to 9, wherein the policy (302) of each topology further comprises a hysteresis parameter, wherein the hysteresis parameter indicates to delay the activation and / or the deactivation of the topology after the activation and / or the deactivation of the topology is triggered.
11. The control entity (300) according to one of the claims 2 to 10, wherein the first threshold of the (n+1)th topology is higher than the first threshold of the nth topology, and / or the second threshold of the (n+1)th topology is higher than the second threshold of the nth topology.
12. The control entity (300) according to one of the claims 2 to 11, wherein the first threshold of the (n+1)th topology is not below the second threshold of the (n+1)th topology.
13. An agent entity (500) for assisting multi-topology routing in a network (400), wherein the agent entity (500) is located at a router (401) of the network (400), and the network (400) comprises a plurality of routers (401), wherein the agent entity (500) is configured to: obtain a set of topologies (301) from a control entity (300), wherein the set of topologies (301) comprises N topologies, wherein an (n+1)th topology of the set of topologies (301) comprises all links of an nth topology of the set of topologies (301), wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; and activate or deactivate a topology of the set of topologies (301) in the router (401) based on an activation indication (303) or a deactivation indication (304) from the control entity (300), or based on a policy (302) for each topology of the set of topologies (301), wherein the policy (302) indicates when to trigger an activation and / or a deactivation of the topology.
14. The agent entity (500) according to claim 13, configured to: receive the activation indication (303) or the deactivation indication (304) from the control entity (300).
15. The agent entity (500) according to claim 13 or 14, configured to: receive the policy (302) for each topology of the set of topologies (301) from the control entity (300).
16. The agent entity (500) according to claim 15, wherein the policy (302) comprises at least one of a first threshold and a second threshold, wherein the first threshold indicates when to trigger an activation of the topology, and the second threshold indicates when to trigger a deactivation of the topology.
17. The agent entity (500) according to claim one of the claims 13 to 16, further configured to: determine whether to activate or deactivate the (n+1)th topology of the set of topologies (301) in the network (400) based on the policy (302) of that topology and an indicator of utilization of links in the network (400).
18. The agent entity (500) according to claims 16 and 17, configured to: determine to activate the (n+1)th topology in the router when the utilization of the links in the network (400) exceeds the first threshold of the (n+1)th topology, and / or determine to deactivate the (n+1)th topology in the router when the utilization of the links in the network (400) is below the second threshold of the (n+1)th topology.
19. The agent entity (500) according to claims 17 or 18, wherein the agent entity (500) is a first agent entity (500) located at a first router (401), and is configured to: send an activation notification or a deactivation notification to a second agent entity (500’) located at a second router (401’) of the plurality of routers (401) in the network (400), wherein the activation notification or the deactivation notification indicates to the second agent entity (500’) that the first agent entity (500) determined to activate or deactivate the (n+1)th topology in the first router (401).
20. The agent entity (500) according to claim 19, configured to: receive an activation response or a deactivation response from the second agent entity (500’), wherein the activation response or the deactivation response indicates to the first agent entity (500) that the second agent entity (500’) determined to activate or deactivate the (n+1)th topology in the second router (401’); and activate or deactivate the (n+1)th topology of the set of topologies (301) in the first router (401).
21. The agent entity (500) according to one of the claims 13 to 20, wherein the policy (302) of each topology further comprises a hysteresis parameter, wherein the hysteresis parameter indicates to delay the activation and / or the deactivation of the topology after the activation and / or the deactivation of the topology is triggered.
22. A method for a control entity (300) managing multi-topology routing in a network (400), wherein the network (400) comprises a plurality of routers (401), and the method comprises: generating a set of topologies (301) for the network (400), wherein the set of topologies (301) comprises N topologies, wherein an (n+1)th topology of the set of topologies (301) comprises all links of an nth topology of the set of topologies (301), wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N;generating a policy (302) for each topology of the set of topologies (301), wherein the policy (302) is for triggering an activation and / or a deactivation of the topology; and configuring each of the plurality of routers (401) with the set of topologies (301).
23. A method for an agent entity (500) assisting multi-topology routing in a network (400), wherein the agent entity (500) is located at a router (401) of the network (400), and the network (400) comprises a plurality of routers (401), wherein the method comprises: obtaining a set of topologies (301) from a control entity (300), wherein the set of topologies (301) comprises N topologies, wherein an (n+1)th topology of the set of topologies (301) comprises all links of an nth topology of the set of topologies (301), wherein N is a positive integer greater than 1, n is a positive integer, and 1 ≤ n < N; and activating or deactivating a topology of the set of topologies (301) in the router based on an activation indication (303) or a deactivation indication (304) from the control entity (300), or based on a policy (302) for each topology of the set of topologies (301), wherein the policy (302) indicates when to trigger an activation and / or a deactivation of the topology.
24. A computer program product comprising a program code for carrying out, when implemented on a processor, the method according to claim 22 or 23.
Citation Information
Patent Citations
Cycle-free multi-topology routing
US20140269413A1
System and method for layer 3 ring protection with adaptive bandwidth microwave links in a network environment
US20140286173A1
