Communication network risk map

Abstract

The disclosed systems, structures, and methods relate to risk assessment of optical networks. An emulation framework includes a risk map engine that includes a performance prediction engine that generates an emulation of an optical network based at least in part on an input network topology and / or service map. The performance prediction engine runs the emulation to predict direct and indirect effects of a risk factor represented in a hypothetical scenario on the optical network based at least in part on received network telemetry data. The risk map engine includes a risk assessment engine that determines a risk associated with the risk factor based at least in part on the predicted direct and indirect effects and a likelihood of the risk factor occurring. The risk assessment engine generates a risk map that displays a combined risk caused by a plurality of risk factors on the optical network.

Description

[0001] Cross-referencing related applications

[0002] This application claims the benefit and priority of U.S. nonprovisional patent application No. 17 / 155,256, entitled "Risk Map for Communication Networks", filed on January 22, 2021. Invention Field

[0003] This invention generally relates to the field of risk assessment of communication networks, and more particularly to systems and methods for performing simulation-based risk assessments of optical communication networks. Background Technology

[0004] Data services are growing exponentially, primarily due to high-bandwidth applications such as cloud computing, video and game streaming, virtual reality, haptics, and many other increasingly popular applications that require ever-increasing amounts of network bandwidth. Any disruption to the underlying transport network of these applications and services could result in significant data loss and consequent revenue loss.

[0005] Furthermore, communication service providers typically specify a minimum guaranteed availability for the provided connectivity in their service level agreements (SLAs), and any breach of this commitment will incur penalties. Network operators need a framework that includes relevant tools for risk assessment and decision-making (taking preventative and proactive measures) to maximize service availability and avoid / minimize such penalties.

[0006] Because numerous complex factors influence the risk assessment process, analytical methods often fail to provide satisfactory solutions. The situation is even more complex when attempting to assess the risk of optical communication networks, such as dense wavelength division multiplexing (DWDM) networks. The optical connections used in these networks are subject to complex interdependencies, also known as channel coupling. Channel coupling in optical connections can be caused by nonlinear effects such as power shift and stimulated Raman scattering (SRS) or cross-phase modulation (XPM). Improving the resilience and survivability of optical networks against potential failures, such as fiber breaks, requires a clear understanding of the complex interdependencies between various network components and services. Current risk assessment models cannot handle these interdependencies. Summary of the Invention

[0007] Advantageously, the present invention provides a system and method for generating risk maps of DWDM networks based on optical behavior simulation. The disclosed simulation platform takes network topology and service graphs as input and employs a series of analytical and machine learning-based optical device models to accurately simulate the behavior of the optical network in possible hypothetical scenarios. This simulation can model both direct and indirect effects, such as power excursion and nonlinear effects (e.g., stimulated Raman scattering, cross-phase modulation) associated with these hypothetical scenarios. Based on the simulation results, risk levels are associated with each risk factor and each service, and a visual risk map is generated to help network operators mitigate risks, thereby improving the reliability and performance of the optical network.

[0008] Previously, implementing the techniques described in this disclosure was impractical because, as described, performing accurate optical behavior simulations of every risk factor across the entire optical network required previously unavailable computational and storage resources. Furthermore, no accurate models of optical network components were previously available. With current advancements in optical component modeling and improvements in the performance and storage capacity of computing platforms, the simulation platform and risk map generation disclosed herein have become commercially viable.

[0009] According to one aspect of the invention, the technology is implemented as a system for risk assessment of an optical network. The system includes a processor, a memory coupled to the processor, and an interface configured to receive network telemetry data from the optical network. A simulation framework resides in the memory and executes on the processor. The simulation framework includes a risk graph engine, which includes a performance prediction engine configured to generate a simulation of the optical network based at least in part on the input network topology. The performance prediction engine is configured to run the simulation to predict, at least in part on the direct and indirect effects of risk factors represented in a hypothetical scenario on the optical network, based on network telemetry data. The risk graph engine also includes a risk assessment engine configured to determine the risk associated with the risk factors based at least in part on the predicted direct and indirect effects of the risk factors and the probability of the risk factors occurring. The risk assessment engine generates a risk graph displaying the combined risk posed by multiple risk factors to the optical network.

[0010] According to some embodiments of the above aspects, the simulation framework also includes a scenario generator that generates hypothetical scenarios for the plurality of risk factors. In these implementations, the performance prediction engine can be configured to run simulations for the hypothetical scenarios generated for each of the plurality of risk factors.

[0011] According to some embodiments of the above aspects, the simulation predicts the direct and indirect impacts on the optical network, at least in part, by predicting the direct and indirect impacts of risk factors on each of a plurality of services on the optical network. The plurality of services can be defined in the input service graph. In such an implementation, the risk assessment engine can generate a risk graph displaying the comprehensive risk posed to each of the plurality of services.

[0012] According to some embodiments of the above aspects, indirect effects may be caused by power offset. According to some embodiments of the above aspects, indirect effects may be caused by optical nonlinear effects, such as stimulated Raman scattering and / or cross-phase modulation. According to some embodiments of the above aspects, the optical network may be a dense wavelength division multiplexing optical network.

[0013] According to some embodiments of the foregoing aspects, the risk graph engine may include a model repository containing models for each type of component used in the optical network. In such an implementation, the performance prediction engine can generate a simulation of the optical network by combining models from the model repository according to the input network topology. According to some embodiments of the foregoing aspects, at least one model in the model repository is a machine learning-based model.

[0014] According to some embodiments of the above aspects, the risk graph engine may include reliability specifications of components of the optical network. In such an implementation, the risk assessment engine can be configured to determine the likelihood of a risk factor occurring based at least in part on the reliability specifications.

[0015] According to some embodiments of the above aspects, the risk assessment engine determines the risk associated with a risk factor based on the predicted residual margin related to the direct and indirect impacts of the risk factor. According to some embodiments of the above aspects, the risk assessment engine determines the risk associated with a risk factor based on predicted interrupted data traffic volume. According to some embodiments of the above aspects, the risk assessment engine determines the risk associated with a risk factor based on predicted revenue loss, which may be determined at least in part based on the service level agreement (SLA) of the services on the optical network. According to some embodiments of the above aspects, the risk assessment engine generates recommendations for mitigating risks on the optical network. According to some embodiments of the above aspects, the risk assessment engine generates a risk map showing the risks constituted by each optical multiplexed segment link in the optical network.

[0016] According to another aspect of the invention, this technology is implemented as a method for risk assessment of an optical network. The method includes: receiving network telemetry data from the optical network on a computer; generating a simulation of the optical network on the computer based at least in part on the input network topology; running the simulation on the computer to predict, at least in part on the network telemetry data, the direct and indirect effects of risk factors, represented as hypothetical scenarios, on the optical network; determining, at least in part on the predicted direct and indirect effects of the risk factors and the probability of the risk factors occurring, the risks associated with the risk factors on the computer; generating and displaying a risk map on the computer showing the combined risks posed by multiple risk factors to the optical network; and mitigating the risks on the optical network based on the risk map.

[0017] According to some embodiments of the above aspects, the method further includes generating hypothetical scenarios for multiple risk factors. In these implementations, running the simulation may include running the simulation separately for each hypothetical scenario generated for each of the multiple risk factors.

[0018] According to some embodiments of the above aspects, running the simulation includes predicting the direct and indirect impacts of risk factors on each of a plurality of services on an optical network. The plurality of services are defined in an input service graph. In such an implementation, generating and displaying the risk graph may include generating and displaying a risk graph illustrating the combined risk posed to each of the plurality of services.

[0019] According to some embodiments of the foregoing aspects, the method further includes generating recommendations for mitigating risks on optical networks. According to some embodiments of the foregoing aspects, identifying risks associated with risk factors includes using service level agreements (SLAs) for services on the optical network to predict revenue loss.

[0020] According to another aspect of the invention, the technology is implemented as a method for generating a risk map of an optical network. The method includes determining multiple risks on a computer. The multiple risks represent the risks posed by multiple risk factors to services on the optical network, and are determined by simulating components of the optical network to determine the direct and indirect effects of the multiple risk factors. The method further includes generating a representation of a comprehensive risk level posed by the multiple risk factors to services on the optical network based on the multiple risks, and displaying the representation of the comprehensive risk level.

[0021] According to some embodiments of the second aspect, the representation of the overall risk level of the generated services includes a list of high-risk services. Displaying the representation of the overall risk level includes displaying the list.

[0022] According to some embodiments of the second aspect, generating a representation of the comprehensive risk level of a service includes mapping the comprehensive risk level to multiple edges in a graph, wherein each edge in the graph represents the comprehensive risk level of the service, and wherein each node in the graph represents an access point of the service. Displaying the representation of the comprehensive risk level includes displaying the graph.

[0023] According to some embodiments of the second aspect, generating a representation of the overall risk level of a service includes generating a list of recommendations for reducing the risk posed to high-risk services. Displaying the representation of the overall risk level includes displaying the list of recommendations.

[0024] According to another aspect of the invention, the technology is implemented as a method for generating a risk map of an optical network. The method includes determining multiple risks on a computer. The multiple risks represent the risk posed by each link in the optical network due to multiple risk factors, which are determined by simulating the components of the optical network to determine the direct and indirect effects of the multiple risk factors. The method further includes generating a representation of a comprehensive risk level based on the multiple risks, and displaying the representation of the comprehensive risk level.

[0025] According to some embodiments of the third aspect, generating a representation of the overall risk level of links in an optical network includes generating a list of optical network elements that pose a high risk to the network. Displaying the representation of the overall risk level includes displaying the list.

[0026] According to some embodiments of the third aspect, generating a representation of the overall risk level of links in an optical network includes mapping the overall risk level to multiple edges in a graph, wherein each edge in the graph represents the overall risk level of links in the optical network, and wherein each node in the graph represents an element of the optical network. Displaying the representation of the overall risk level includes displaying the graph.

[0027] According to some embodiments of the third aspect, generating a representation of the overall risk level of the links in the optical network includes generating a list of recommendations for reducing the risk of high-risk links. Displaying the representation of the overall risk level includes displaying the list of recommendations. Attached Figure Description

[0028] The features and advantages of this disclosure will become apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0029] Figure 1 A simplified example diagram of an optical network is shown to illustrate the impact of a fiber breakage incident on several connections;

[0030] Figure 2 An example of a risk level generated by a combination of probability and impact is shown;

[0031] Figure 3 It is a block diagram of a simulation platform based on the implementation method of the disclosed technology;

[0032] Figure 4 An example of a risk table that can be generated by a risk graph engine based on an implementation of the disclosed technology is shown;

[0033] Figure 5 An example of a risk graph that can be generated by a risk graph engine based on an implementation of the disclosed technology is shown, which displays the risk posed to each service and the risk posed to each link;

[0034] Figure 6 It is a block diagram of a computer system that can be used to perform simulations based on the implementation of the disclosed technology.

[0035] It should be understood that similar features are identified by similar reference numerals throughout all the drawings and corresponding descriptions. Furthermore, it should be understood that the drawings and the following description are for illustrative purposes only, and this disclosure is not intended to limit the scope of the claims. Detailed Implementation

[0036] Various representative embodiments of the disclosed technology will be described more fully below with reference to the accompanying drawings. However, the technical concept can be embodied in many different forms and should not be construed as limited to the representative embodiments described herein. In the drawings, the dimensions and relative dimensions of layers and regions may be exaggerated for clarity. Throughout the specification, similar figures refer to similar elements.

[0037] It should be understood that although the terms first, second, third, etc., may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used to distinguish one element from another. Therefore, without departing from the teachings of this disclosure, the first element discussed below may be referred to as the second element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0038] It should be understood that when one element is referred to as "connected" or "coupled" to another element, it can be directly connected or coupled to the other element, or there may be an intermediate element. Conversely, when one element is referred to as "directly connected" or "directly coupled" to another element, there is no intermediate element. Other terms used to describe the relationship between elements should be interpreted in a similar manner (e.g., "between" vs. "directly between," "adjacent" vs. "directly adjacent," etc.). Furthermore, it should be understood that elements can be "coupled" or "connected" by mechanical, electrical, communicative, wireless, optical, or other means, depending on the type and nature of the elements being coupled or connected.

[0039] The terminology used herein is for describing specific, representative embodiments only and is not intended to limit the technology. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” used herein also include the plural forms. It should also be understood that the term “comprising” as used herein is used to indicate the presence of the described features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0040] The functions of the various elements shown in the figure, including any functional blocks labeled "processor," can be provided by using dedicated hardware and hardware capable of executing instructions, associated with appropriate software instructions. When provided by a processor, these functions can be provided by a single dedicated processor, a single shared processor, or multiple separate processors, some of which may share resources. In some implementations of this technology, the processor can be a general-purpose processor, such as a central processing unit (CPU), or a processor dedicated to a specific purpose, such as a digital signal processor (DSP). Furthermore, the explicit use of the term "processor" should not be construed as referring specifically to hardware capable of executing software, but may implicitly include, but is not limited to, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), read-only memory (ROM) for storing software, random access memory (RAM), and non-volatile storage devices. Other conventional and / or custom hardware may also be included.

[0041] A software module, or simply, implied as a module or unit of software, may herein be represented as a flowchart element or any combination of other elements indicating the execution of process steps and / or textual descriptions. Such modules may be executed by hardware, whether explicitly shown or implicitly shown. Furthermore, it should be understood that a module may include, for example, but not limited to, computer program logic, computer program instructions, software, stacks, firmware, hardware circuitry, or combinations thereof, providing the required capabilities. It should also be understood that a “module” generally defines a logical grouping or organization of related software code or other elements associated with the defined functionality as described above. Therefore, those skilled in the art will understand that in some implementations, specific code or elements described as part of a “module” may be placed in other modules, depending on the logical organization of the software code or other elements, and such modifications are within the scope of this disclosure as defined in the claims.

[0042] It should also be noted that, as used in this article, the term "optimization" means improvement. It is not used to convey that the technique has produced an objectively "best" solution, but rather an improved solution (at least in one aspect). In the context of memory access, it generally means that the efficiency or speed of memory access can be improved.

[0043] As used herein, the term "determine" generally means to perform a direct or indirect operation, calculation, decision, lookup, measurement, or test. In some cases, such determination may be approximate. Therefore, determining a value indicates that the value or an approximation of the value was obtained through direct or indirect operations, calculations, decisions, lookups, measurements, tests, etc. If an item is "predetermined," it is determined at any time prior to the moment it is indicated as "predetermined."

[0044] This technology can be implemented as a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium (or medium) storing computer-readable program instructions that, when executed by a processor, cause the processor to perform various aspects of the disclosed technology. A computer-readable storage medium may be, for example, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of these devices. A non-exhaustive list of more specific examples of computer-readable storage media includes: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), flash memory, optical disks, memory sticks, floppy disks, mechanical or visual encoding media (e.g., punched cards or barcodes), and / or any combination of these media. As used herein, computer-readable storage media should be construed as non-transient computer-readable media. It should not be construed as transient signals, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0045] It should be understood that computer-readable program instructions can be downloaded from a computer-readable storage medium to the corresponding computing or processing device, or downloaded to an external computer or external storage device via a network such as the Internet, local area network, wide area network, and / or wireless network. A network interface in each computing / processing device can receive computer-readable program instructions over the network and forward these instructions for storage in a computer-readable storage medium within the corresponding computing or processing device. Computer-readable program instructions used to perform operations of this disclosure can be assembly instructions, machine instructions, firmware instructions, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages.

[0046] All descriptions and specific examples of the principles, aspects, and implementations of this technology herein are intended to include their structural and functional equivalents, whether they are currently known or will be developed in the future. Therefore, for example, those skilled in the art will understand that any block diagram herein represents a conceptual view of an illustrative circuit embodying the principles of this technology. Similarly, it should be understood that any flowchart, diagram, state transition diagram, pseudocode, etc., represents various processes that can be represented substantially by computer-readable program instructions. These computer-readable program instructions can be provided to a processor or other programmable data processing apparatus to create a machine such that the instructions, executable by a computer processor or other programmable data processing apparatus, create methods for implementing the functions / actions specified in the flowcharts and / or block diagram blocks. These computer-readable program instructions can also be stored in a computer-readable storage medium that can instruct a computer, programmable data processing apparatus, and / or other device to operate in a particular manner, such that the computer-readable storage medium having instructions stored therein includes an article of art containing instructions that implement various aspects of the functions / actions specified in the flowcharts, diagrams, state transition diagrams, pseudocode, etc.

[0047] Computer-readable program instructions can also be loaded onto a computer, other programmable data processing apparatus or other equipment to cause a series of operational steps to be performed on the computer, other programmable apparatus or other equipment to produce a process implemented by the computer, thereby enabling the instructions to be executed on the computer, other programmable apparatus or other equipment to perform the functions / actions specified in flowcharts, flowcharts, state transition diagrams, pseudocode, etc.

[0048] In some alternative implementations, the functions marked in flowcharts, diagrams, state transition diagrams, pseudocode, etc., may not be executed in the order indicated in the diagram. For example, two boxes shown consecutively in a flowchart may actually be executed almost simultaneously, or sometimes in reverse order, depending on the functions involved. It should also be noted that each function marked in the diagram, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified function or action, or by a combination of dedicated hardware and computer instructions.

[0049] Based on these fundamental principles, several non-limiting examples will now be considered to illustrate various implementations of various aspects of this disclosure. It should be noted that although the various implementations are described in conjunction with DWDM networks, it will be apparent to those skilled in the art that similar techniques can be used for other types of communication networks.

[0050] According to the disclosed technology, one approach to obtaining a comprehensive risk map of a DWDM network is based on optical behavior simulation. To facilitate such simulation, a simulation platform can be used that takes the network topology and service graph as input, employing various analytical and machine learning-based optical device models to accurately simulate the behavior of the optical network in possible "hypothetical" scenarios. In one implementation, the simulation platform can use information about margin levels and built-in capabilities for optical performance prediction to assess the risk of service failures.

[0051] Risk factors in optical DWDM networks fall into three main categories: fiber optic link failures, primarily caused by fiber breaks, but in some cases by equipment malfunctions and human error; optical component failures; and miscellaneous events that could lead to disruptions, including human error. Each of these factors can have both direct and indirect impacts.

[0052] For example, if a fiber breakage occurs, the entire optical multiplex section (OMS) (i.e., the section between two optical add-drop multiplexers) will fail. This failure has both direct and indirect effects. The direct effects will immediately appear on all optical channel (OCh) routes passing through the failed OMS. The indirect effects are caused by, for example, changes in channel load and the resulting power shifts (e.g., due to complex fiber amplifier behavior), which may affect many other OCh services throughout the network.

[0053] Figure 1 An event is illustrated. In optical network 100, the fiber optic link 102 between reconfigurable optical add-drop multiplexers (ROADMs) A130 and ROADM B 132 (along which a number of optical amplifiers 104 exist) is severed (shown as fiber break 106). As a direct effect, this will cause all OCh routes passing through this link to be dropped, including connection ABDG 160 (i.e., the connection passing through ROADM A 130, ROADM B 132, ROADM D 136, and ROADM G 142). However, this is not the only impact of this event on optical network 100, as there are many interdependencies between optical channels that can lead to indirect effects, even affecting OCh routes that do not pass through fiber link 102.

[0054] For example, the connection CBDE 162 (via ROADM C 134, ROADM B 132, ROADM D 136, and ROADM E138) will be indirectly affected because the channel load on the fiber link between ROADM B 132 and ROADM D 136 will change. When fiber link 102 is disconnected, there will be no power on the wavelength carrying connection ABDG 160. This will affect the power levels on other channels between ROADM B 132 and ROADM D 136, including connection CBDE 162. These other channels carry different wavelengths in the DWDM network. In some DWDM networks, there may be, for example, 80 channels in the C-band (1530 nm–1565 nm), each carrying a different wavelength on a single fiber—though there may be more or fewer channels in the C-band. Some networks are also able to add additional channels on a single fiber using wavelengths in the L-band (1565 nm–1625 nm).

[0055] As power levels change, connection performance may vary, as indicated by changes in the destination bit error rate. When connection performance falls below a threshold, the connection may be lost. Therefore, even if fiber breakage 106 does not directly affect connection CBDE 162, there are indirect effects that could lead to connection loss, increasing the risk of connection loss for CBDE 162.

[0056] The third connection, CDE 164 (i.e., the connection via ROADM C 134, ROADM D 136, and ROADM E 138), is also indirectly affected by fiber break 106. This is because the power shift on connection CBDE 162, discussed above, also affects the channel between ROADM D 136 and ROADM E 138. Although connection CDE 164 does not share any fiber link with the original connection ABDG160 where fiber break 106 occurred, it is still affected because the power level on connection CBDE 162 changes. Therefore, there is a second-level indirect impact on connection CDE 164, which can lead to connection loss and thus increase the risk of connection CDE 164 loss.

[0057] The fourth connection, AEF 166 (i.e., the connection via ROADM A.130, ROADM E 138, and ROADM F 140), is not directly or indirectly affected by the fiber break 106. This is because there is no fiber optic link in connection AEF 166 shared with any directly or indirectly affected connection.

[0058] Information regarding the risks of such direct and indirect impacts is of significant value to network operators, as any type of disruption to the transmission network can result in lost revenue. Network operators' customers typically have service level agreements (SLAs), which usually impose financial penalties for failure to meet specified requirements. The risk assessment framework of the type described in this article allows network operators to better understand the direct and indirect risks associated with various possible scenarios, thereby enhancing their ability to take preventative measures to reduce or mitigate risks and / or lost revenue.

[0059] For DWDM networks, indirect effects such as those described above are typically caused by power shifts or nonlinear effects. Power shifts can be caused by power adjustments made by embedded power control and adjustment algorithms within the optical network, which may occur when channels are added or removed. Due to these power adjustments, adding or removing an optical channel can affect all other channels on the fiber. Power shift effects can also be caused by the complex behavior of optical amplifiers (such as erbium-doped fiber amplifiers (EDFAs) commonly used in DWDM networks). In response to changes in channel load and power levels on the fiber link, an EDFA may cause power shifts in all channels of that fiber link.

[0060] Besides power shift, indirect effects can also be caused by nonlinear effects such as stimulated Raman scattering (SRS) or cross-phase modulation (XPM). SRS causes power transfer from a low-wavelength channel to a high-wavelength channel during bipolar transmission through an optical fiber, which can occur when the EDFA attempts to compensate for power balance in the fiber. In the C+L band, SRS crosstalk is approximately proportional to the frequency difference between the two channels. It is important to note that the gain from the SRS effect in the C+L band is much higher (up to 3 times in some cases) compared to C-band or L-band-only operation. XPM is a nonlinear effect caused by a change in one wavelength in the fiber, which alters the phase of another wavelength in the fiber through the optical Kerr effect.

[0061] The sources of these indirect effects in DWDM networks are well documented in the literature and are known to those skilled in the art. These effects can also be simulated or modeled, as described below.

[0062] For reference Figure 2This describes the concept of "risk" as used in this disclosure. In the context of the disclosed technology, risk can be viewed as the expected value of an undesirable outcome. Risk combines the probabilities of various possible events with some measure of the corresponding loss into a single value. This can be roughly expressed as:

[0063] Risk = Probability of the risk factor × Expected loss caused by the risk factor

[0064] Figure 2 Matrix 200 in the diagram shows an example of risk levels resulting from the combination of probability (likelihood) and loss (impact). The vertical axis 202 represents the likelihood of a risk factor occurring, ranging from "very low probability" to "very high probability." The horizontal axis 204 shows the impact caused by that risk factor, ranging from "negligible" to "severe." It can be seen that if the impact of an event is severe, then even an event with a very low probability may present a "moderate" risk.

[0065] It should be understood that Matrix 200 represents only one example of risk determination based on probability and impact. Different risk levels can be assigned to any given probability and impact, depending on the network operator, service, type of risk factor being assessed, etc. It should also be understood that other risk determination methods may exist that can be used according to this technique. Figure 2 This is only used to illustrate the general concept of risk.

[0066] Based on the above discussion of risks, direct and indirect impacts, and the potential financial consequences for network operators if SLA requirements are not met, it would be beneficial to provide network operators with tools to understand and mitigate these risks. These tools help identify the direct and indirect impacts of optical network failures and inform network operators for redundancy planning and / or preventative action. These tools can quantify the aggregate risk of each existing connection and provide a list or diagram of the most vulnerable services operating on the network. Furthermore, these tools can quantify the risk posed by each identified risk factor (e.g., link failure) and identify critical links and / or components in the network. This information can be used to recommend modifications to service provisioning and / or potential redundancy to mitigate risks.

[0067] According to the disclosed technology, a simulation framework can be used to better understand risk. This framework collects telemetry data from the optical network and uses optical device models to simulate the network's behavior in various hypothetical scenarios. This simulation framework can be used to assess the availability risk of each optical connection on the network in the event of any OMS link failure and to identify critical OMS links based on the risk posed by OMS link failure to existing connections. The framework can also identify critical wavelengths based on the risk caused by wavelength loss. Based on the simulation results, a risk map of the entire network can be generated and visualized to provide risk information for all operating services based on their respective margin levels. The simulation framework can also recommend contingency plans and modifications to data rates or routing and / or wavelength allocation for existing connections to increase availability margins and / or reduce associated risks for specific connections or the entire network.

[0068] Figure 3 A structural diagram of this simulation framework 300 is shown. This framework includes a risk graph engine 302, which takes network topology 304, service graph 306, "hypothetical" scenarios 308, and network telemetry data 310 as input. As will be described more fully below, the risk graph engine 302 uses this information to simulate the optical network under a series of "hypothetical" scenarios 308 and generates a risk graph 312 as output. The risk graph engine 302 includes a model repository 314, a performance prediction engine 316, a reliability specification 318, and a risk assessment engine 320.

[0069] Network topology 304 includes information about the devices, components, and links that make up the optical network being risk-assessed. These devices and components may include, for example, hardware (such as boards) on each network node, EDFAs, ROADMs, repeaters, etc. Specification information about these devices, components, and links may also be included in network topology 304.

[0070] Service Diagram 306 includes information about services used on the network. Service Diagram 306 may include the source, destination, bandwidth, SLA requirements, and penalties of the services.

[0071] "Hypothetical" scenario 308 is a specific scenario that the risk graph engine 302 will simulate. For example, such a "hypothetical" scenario could include a fiber break in a specific fiber connection, a failure of a specific optical amplifier, a hardware failure of a specific component, channel deletion, etc. The "hypothetical" scenarios are generated by scenario generator 322, which generates scenarios one by one for all known or anticipated risks and sends each such scenario to the risk graph engine 302. In some implementations, each "hypothetical" scenario 308 represents a single known or anticipated risk.

[0072] Network telemetry data 310 includes actual network performance information. To obtain network telemetry data 310, sensors can be used to monitor the performance of OMS links, channels, and / or network components. This provides the risk graph engine 302 with information about parameters such as power levels, connectivity performance, optical signal-to-noise ratio (OSNR), and bit error rate on channels and / or OMS links. In some implementations, this information can be provided to the risk graph engine 302 in real-time or near real-time.

[0073] Within the Risk Graph Engine 302, the Model Repository 314 is a database storing models for each class of components used in the network. These models can be analytical—that is, based on known algorithms or heuristics that simulate specific types of components. For some types of components, such as EDFA, component performance can be complex, so known analytical models may not achieve the required accuracy in simulating component behavior. To handle these types of components in the network, models can be based on known machine learning (ML) techniques rather than on known equations governing the physical properties of the components. ML-based models are typically trained using real-world behavioral data of the types of devices being used. ML techniques that can be used to model or simulate the behavior of various optical components can include neural networks (e.g., deep neural networks, convolutional neural networks, or other currently known or later developed types of neural networks), regression, decision trees, Bayesian machine learning techniques, K-nearest neighbor techniques, random forest techniques, and / or other known or later developed machine learning techniques, including supervised learning, unsupervised learning, and / or reinforcement learning.

[0074] In some implementations, training of the machine learning-based model is performed "offline" on different systems, and the machine learning-based models available in the model repository 314 are pre-trained. In other implementations, at least a portion of the model can be trained on the same system used to implement the risk graph engine 302 using, for example, network telemetry data 310.

[0075] The performance prediction engine 316 generates a simulation of the entire network by combining models from model repository 314 according to network topology 304. The network simulation uses detailed individual models from model repository 314 to simulate the direct and indirect effects of risk factors on the network, such as power shift and nonlinear effects. This allows for the simulation of complex nonlinear and power shift-related interdependencies in optical connections, enabling a better assessment of service-related risks.

[0076] The performance prediction engine 316 runs a separate simulation session for each "hypothetical" scenario 308, which predicts the impact on all services on the network (determined by the service graph 306), including both direct and indirect impacts. Network telemetry data 310 is also used in the performance prediction engine 316 to provide more accurate information about the actual network, such as OSNR and power levels on channels.

[0077] Performance prediction engine 316 estimates a metric for the quality of transmission (QoT) of each OCh (i.e., each service) for each risk factor. In some implementations, this QoT estimate can be represented as an OSNR loss (in dB) per OCh caused by the occurrence of the risk factor (i.e., the "hypothetical" scenario). This OSNR loss can then be used to calculate a margin of safety based on the current margin of the service minus the estimated OSNR loss (also known as an "OSNR penalty"). The margin of safety for each service and for each risk factor is then sent to risk assessment engine 320. In some implementations, performance prediction engine 316 may also determine a confidence interval for each predicted margin of safety to specify the accuracy of the simulation. This confidence interval can be determined based on the accuracy and / or confidence interval of each model and / or simulator combined to simulate network performance.

[0078] In some implementations, services potentially affected by failures have recovery mechanisms to provide protection or restoration. These mechanisms may be triggered by failures and may result in various recovery or reconfiguration actions, which can have direct and indirect impacts. Therefore, in some implementations, a two-round impact assessment and risk evaluation will be conducted. In the first round, the performance prediction engine 316 runs simulations for each risk factor to estimate the performance of each service before any recovery and / or reconfiguration action. In the second round, after the recovery and / or reconfiguration action has occurred, the performance prediction engine 316 estimates the performance of each service under each risk factor. This allows for the evaluation of the effectiveness of such protection or recovery mechanisms and an understanding of their direct and indirect impacts on the network.

[0079] It should be understood that the performance prediction engine 316 runs many computationally intensive simulations. In some implementations, these simulations can be executed on a single system, which may include multiple processors. In some implementations, the simulations can be distributed across many computers on a network. In some implementations, the model can be contained within the optical components being modeled, thus the simulation is distributed, allowing components in the network to execute the model for simulation.

[0080] Reliability Specification 318 includes a database of reliability information for the optical components that make up the network. This information may be in the form of parameters, such as the mean time between failures (MBTF) or mean time to repair (MTTR) of the components used in the network. This information is typically published by the optical network equipment manufacturer. It should be understood that in some embodiments, other reliability metrics may be used. Furthermore, for some risk factors, such as those that depend on the behavior of the network operator rather than equipment failure, reliability metrics such as MBTF or MTTR may not be available or suitable as measures of the probability of occurrence of the risk factor, and therefore other metrics may be used. Generally, the goal of Reliability Specification 318 is to provide information about the probability of occurrence of various risk factors simulated by the system.

[0081] Once the expected performance prediction engine 316 has simulated the impact of each "hypothetical" scenario on each service, information about these impacts is sent to the risk assessment engine 320. The risk assessment engine 320 takes these impacts (e.g., margin of safety), along with the SLA requirements for different services (from service diagram 306) and the probability of occurrence of all risk factors (based on reliability specification 318) as input, and calculates the overall risk for each service and the risk borne by each risk factor. This can be achieved, for example, by combining the above... Figure 2 The description method combines the severity of the impact (based on residual margin) with the probability of the risk factor occurring to determine the risk level. Once this is done, for each service, the risk levels associated with each risk factor can be combined to provide a comprehensive risk for the service. Similarly, for each risk factor, the risk levels for each service can be combined to provide the overall risk constituted by each risk factor. In some implementations, the risk assessment engine 320 can also generate an overall network risk score for the network. In some implementations, MTTR information from reliability specification 318 can also be used, for example, to determine the predicted outage duration.

[0082] The risk assessment engine 320 also generates a risk map 312 and a list of recommendations for reducing or mitigating the risk of each high-risk connection or the entire network. As discussed in more detail below, in some implementations, the risk map 312 may take one or more visual forms, such as a table, an OCh risk map showing the risk posed to each service, and / or a critical OMS link map showing the risk posed to each link in the network. Recommendations or contingency plans may include modifying service data rates or routing and / or wavelength allocation to increase availability margins. Depending on the SLA, in some cases, such as when the predicted risk is low or the predicted service outage duration is short, the recommendation may be to do nothing.

[0083] In some implementations, the risk assessment engine 320 may also generate other representations of risk, such as a list of optical network elements (e.g., links or components) identified by the risk graph engine as posing a high risk to the network. In some implementations, the risk assessment engine 320 may generate a list of high-risk connections or services. For example, whether a network element poses a high risk to the network, or whether a connection or service is a high-risk connection or service, can be determined by comparing the risk to a threshold.

[0084] The risk assessment engine 320 can provide appropriate risk metrics suitable for different scenarios and / or network operators using different measures of expected impact or loss. In some implementations, risk can be determined using the expected loss of service due to connection interruption. To use this risk metric, the risk assessment engine 320 can estimate that connection availability will be lost if the remaining margin falls below a threshold (e.g., 1 dB). The risk posed by each risk factor can then be calculated using the amount of interrupted data service for each connection (based on the estimated loss of connection availability determined by the remaining margin, and, for example, the amount of data service from network telemetry data 310) and the probability of the corresponding risk factor. Using this risk metric, high-risk services and critical risk factors can be identified, and recommendations and contingency plans for minimizing service interruptions can be provided.

[0085] In some implementations, expected revenue loss based on the SLA can be used as a measure of risk. To use this risk measure, the Risk Assessment Engine 320 can estimate the revenue loss that each risk channel will cause based on the projected residual margin and the SLA requirements for each service. Risk is then calculated by combining the projected monetary loss with the probability of the corresponding risk factor. Using this risk measure, risk assessments, risk maps, and recommendations can focus on the potential penalties of service interruption.

[0086] It should be understood that other risk metrics can also be used. For example, Risk Assessment Engine 320 can assess risk based on, for example, the predicted time to repair or restore services or other metrics. It should also be understood that in some implementations, the risk metrics used can be selected and changed by the system user.

[0087] For reference Figure 4This describes an example of a risk table 400 that can be generated by the risk graph engine 302. In risk table 400, columns represent risk factors 402, with each column representing a single risk factor that may pose a risk to the availability of network services. Rows in risk table 400 represent services 404, with each row representing an available OCh service on the network. Therefore, each cell in the table represents the predicted expected loss or impact (e.g., predicted residual margin) on an OCh due to the direct and indirect effects of a risk factor. Depending on the risk metric used, these cells may display other values, such as predicted loss of gains. Thus, risk table 400 provides a “big picture” of optical network vulnerabilities. It should be understood that any values ​​shown in risk table 400 are for illustrative purposes only.

[0088] As can be seen from Risk Table 400, risk factors can be grouped into sets of columns, each set representing similar risk factors. Therefore, the "Fiber Link Disconnection" set 406 includes columns representing the risks resulting from the disconnection of fiber links on each OMS link in the network. The "Component Failure" set 408 includes columns for each critical board or component that may fail. The "Miscellaneous Events" set 410 includes columns for each miscellaneous event, including, for example, channel addition or removal operations performed by the network operator, which may pose a risk to services on the network.

[0089] Table 400 also includes: row 412, which displays the composite risk constituted by each risk factor; column 414, which displays the composite risk for each service; and cell 416, which displays the network risk score. It should be understood that, in various implementations, the data shown in risk table 400 can be organized or visualized in different ways. For example, some implementations may use color coding in the cells of table 400 based on, for example, residual margin, so that color provides an overview even in a very narrowed view. In some implementations, table 400 may not be provided to the user in a visual form, but may be used internally to calculate, for example, composite risk, which can be visualized using other types of visualization (e.g., risk graphs).

[0090] Figure 5The diagram illustrates the OCh risk graph 502 and the critical OMS link risk graph 504. OCh risk graph 502 visually represents the overall risk of each service, while critical OMS link risk graph 504 visually represents the risk posed by each link. OCh risk graph 502 shows a service graph represented graphically, where nodes 520 represent access points where services are provided to clients, and edges 522 represent services. It can be seen that many services may exist between two access points, therefore multiple edges 522 may exist between two nodes 520. Each edge 522 can use representations such as color or line style to indicate the overall risk of the service represented by the edge 522. It can be seen that two edges 522 between the same nodes 520 can be associated with different overall risk levels. For example, this could be due to differences in the underlying physical routing of the services, differences in the SLAs associated with the services, etc.

[0091] The critical OMS link risk diagram 504 is based on OMS links in the network topology. Each node in 540 represents a ROADM or other element of the optical network, and each edge in 542 represents an OMS link between two ROADMs or other elements. Each edge 542 uses a visual representation, such as color or line style, to indicate the risk posed by the OMS link represented by edge 542.

[0092] It should be noted that since access points (i.e., nodes 520) in OCh risk diagram 502 are typically associated with routers co-located with optical nodes (e.g., ROADMs), there may be a correspondence between nodes 520 in OCh risk diagram 502 and some nodes 540 (i.e., ROADMs) in critical OMS link risk diagram 504. Figure 5 This correspondence is illustrated in the example visualization shown. It should be understood that OCh risk diagram 502 and critical OMS link risk diagram 504 can be displayed together, as shown... Figure 5 As shown, or they may be shown individually. It should also be understood that, for illustrative purposes, Figure 5 The risk diagrams shown depict optical networks far less complex than actual optical networks using the techniques described in this disclosure. For optical networks of "normal" complexity, risk diagram visualization may allow zooming and panning to view the entire risk diagram.

[0093] Figure 6A computer system 600 is shown that can be used, for example, to execute the simulation framework described above. Such a computer system 600 can receive real-time or near-real-time network telemetry data from the optical network, as well as other information used in the simulation platform, such as network topology information, service graph information, and "hypothetical" scenario information or risk factors. The computer system 600 can then execute the simulation framework to generate risk maps, recommendations, and contingency plans that can be used to improve the reliability and performance of the optical network.

[0094] Computer system 600 can be a multi-user computer, a single-user computer, a server, an embedded control system, a computer providing services in the "cloud," or any other computer system currently known or developed hereafter. Furthermore, it should be recognized that some or all components of computer system 600 can be virtualized. For example... Figure 6 As shown, computer system 600 includes one or more processors 602, memory 610, storage interface 620, and network interface 640. These system components are interconnected via bus 650, which may include one or more internal and / or external buses (not shown) (e.g., PCI bus, Universal Serial Bus, IEEE 1394 FireWire bus, SCSI bus, Serial-ATA bus, etc.), to which various hardware components are electrically coupled.

[0095] Memory 610 may be random access memory or any other type of memory, and may contain data 612, operating system 614, and program 616. Data 612 may be any data that serves as input or output to any program in computer system 600. Operating system 614 may be an operating system such as Microsoft Windows or Linux. Program 616 may be any program or set of programs that includes program instructions, wherein these program instructions can be executed by a processor to control the actions taken by computer system 600.

[0096] Storage interface 620 is used to connect a storage device (e.g., storage device 625) to computer system 600. One type of storage device 625 is a solid-state drive, which can use integrated circuit components to persistently store data. Different types of storage devices 625 are hard disk drives, such as electromechanical devices that use magnetic storage devices to store and retrieve digital data. Similarly, storage device 625 can be an optical drive, a card reader that receives removable memory cards (e.g., SD cards), or a flash memory device that can be connected to computer system 600 via, for example, a universal serial bus (USB).

[0097] In some implementations, computer system 600 may use well-known virtual memory techniques that make programs of computer system 600 behave as if they could access a large contiguous address space, rather than multiple smaller memory spaces, such as memory 610 and storage device 625. Therefore, although data 612, operating system 614, and program 616 appear to reside in memory 610, those skilled in the art will recognize that these items are not necessarily all contained in memory 610 simultaneously.

[0098] Processor 602 may include one or more microprocessors and / or other integrated circuits. Processor 602 executes program instructions stored in memory 610. When computer system 600 starts, processor 602 may first execute the boot routine and / or program instructions that make up operating system 614.

[0099] Network interface 640 is used to connect computer system 600 to other computer systems or networked devices (not shown) via network 660. Network interface 640 may include a combination of hardware and software that allows communication via network 660. In some implementations, network interface 640 may be a wireless network interface. The software in network interface 640 may include software that communicates via network 660 using one or more network protocols. For example, network protocols may include Transmission Control Protocol / Internet Protocol (TCP / IP). In some implementations, computer system 600 may receive network telemetry data via network interface 640. In some implementations, network telemetry data may be received via other input / output interfaces (not shown), such as a USB connection or other known interfaces.

[0100] It should be understood that Computer System 600 is merely an example, and the disclosed techniques can be used with computer systems or other computing devices with different configurations.

[0101] It should be understood that although the embodiments presented herein have been described in conjunction with specific features and structures, various modifications and combinations can be made without departing from these disclosures. Therefore, the specification and drawings should be considered merely as illustrations of the implementations or embodiments discussed and the principles defined in the appended claims, and are intended to cover any and all modifications, variations, combinations, or equivalents falling within the scope of this disclosure.

Claims

1. A system for risk assessment of optical networks, the system comprising: processor; A memory coupled to the processor; An interface configured to receive network telemetry data from the optical network; as well as A simulation framework, residing in the memory and executing on the processor, includes a risk graph engine comprising: A performance prediction engine is configured to generate a simulation of the optical network based at least in part on the input network topology, the performance prediction engine is configured to run the simulation to predict, at least in part on the network telemetry data, the direct and indirect impacts of risk factors represented in a hypothetical scenario on the optical network, determine the remaining margin of the service level based on the risk factors and the current margin of the service, and determine a confidence interval for each determined remaining margin. as well as A risk assessment engine is configured to determine the risk associated with the risk factor based at least in part on the predicted direct and indirect effects of the risk factor and the probability of the risk factor occurring. The risk assessment engine generates a risk map showing the combined risk posed by multiple risk factors to the optical network.

2. The system according to claim 1, wherein, The simulation framework also includes a scenario generator that generates hypothetical scenarios for the plurality of risk factors, wherein the performance prediction engine is configured to run the simulation for the hypothetical scenarios generated for each of the plurality of risk factors.

3. The system according to claim 1 or 2, wherein, The simulation predicts the direct and indirect impacts on the optical network, at least in part, by predicting the direct and indirect impacts of the risk factors on each of a plurality of services defined in an input service graph, wherein the risk assessment engine generates a risk graph that displays the combined risk posed to each of the plurality of services.

4. The system according to claim 1 or 2, wherein, The indirect effects are caused by power offset.

5. The system according to claim 1 or 2, wherein, The indirect effects are caused by optical nonlinear effects.

6. The system according to claim 1 or 2, wherein, The optical network includes a dense wavelength division multiplexing optical network.

7. The system according to claim 1 or 2, wherein, The risk graph engine also includes a model repository containing models for each type of component used in the optical network, and wherein the performance prediction engine generates a simulation of the optical network by combining the models in the model repository according to the input network topology.

8. The system according to claim 7, wherein, At least one model in the model repository is a machine learning-based model.

9. The system according to claim 1 or 2, wherein, The risk graph engine also includes reliability specifications for components of the optical network, and the risk assessment engine is configured to determine the likelihood of the risk factor occurring based at least in part on the reliability specifications.

10. The system according to claim 1 or 2, wherein, The risk assessment engine determines the risk associated with the risk factor based on the residual margin related to the direct and indirect effects of the risk factor.

11. The system according to claim 1 or 2, wherein, The risk assessment engine determines the risks associated with the risk factors based on the predicted data interruption volume.

12. The system according to claim 1 or 2, wherein, The risk assessment engine determines the risks associated with the risk factors based on predicted gains or losses.

13. The system according to claim 12, wherein, The predicted revenue loss is determined at least in part based on the service level agreement (SLA) of the services on the optical network.

14. The system according to claim 1 or 2, wherein, The risk assessment engine also generates recommendations on reducing risks on the optical network.

15. The system according to claim 1 or 2, wherein, The risk assessment engine generates a risk map that displays the risks associated with each optical multiplexed segment link in the optical network.

16. A method for risk assessment of optical networks, the method comprising: Receive network telemetry data from the optical network on the computer; A simulation of the optical network is generated on the computer, at least in part, based on the input network topology; The simulation is run on the computer to predict, at least in part, the direct and indirect impacts of risk factors, represented as hypothetical scenarios, on the optical network, based on the network telemetry data. The remaining margin of the service level is determined based on the risk factors and the current margin of the service, and a confidence interval is determined for each determined remaining margin; The risk associated with the risk factor is determined on the computer, based at least in part on the predicted direct and indirect effects of the risk factor and the probability of the risk factor occurring. A risk map, showing the combined risk posed by multiple risk factors to the optical network, is generated and displayed on the computer. as well as Risks on the optical network are reduced based on the risk map.

17. The method of claim 16, further comprising generating hypothetical scenarios for the plurality of risk factors, wherein, Running the simulation involves running the simulation individually for each of the multiple risk factors, based on the hypothetical scenarios generated for each of the risk factors.

18. The method according to claim 16 or 17, wherein, Running the simulation includes predicting the direct and indirect impacts of the risk factors on each of a plurality of services on the optical network, the plurality of services being defined in an input service graph, and wherein generating and displaying the risk graph includes generating and displaying a risk graph showing the combined risk to each of the plurality of services.

19. The method of claim 16 or 17, further comprising generating recommendations on reducing risks on the optical network.

20. The method according to claim 16 or 17, wherein, Identifying the risks associated with the aforementioned risk factors includes using the service level agreements (SLAs) of services on the optical network to predict revenue loss.

21. A method for generating a risk map of an optical network, the method comprising: Multiple risks are identified on a computer, representing multiple risk factors that pose a risk to services on the optical network. These multiple risks are determined by simulating components of the optical network to determine the direct and indirect effects of the multiple risk factors. The remaining margin of the service level is determined based on the risk factors and the current margin of the service, and a confidence interval is determined for each determined remaining margin; Based on the aforementioned multiple risks, a comprehensive risk level representation is generated for the services on the optical network caused by the aforementioned multiple risk factors. as well as This displays a representation of the overall risk level.

22. The method according to claim 21, wherein, The representation of the overall risk level of the generated services includes generating a list of high-risk services, and displaying the representation of the overall risk level includes displaying the list.

23. The method according to claim 21 or 22, wherein, The representation of the comprehensive risk level of the service includes mapping the comprehensive risk level to a plurality of edges in a graph, wherein each edge in the graph represents the comprehensive risk level of the service, and wherein each node in the graph represents an access point of the service; and wherein displaying the representation of the comprehensive risk level includes displaying the graph.

24. The method according to claim 21 or 22, wherein, The representation of the overall risk level of the service includes generating a list of recommendations for reducing the risk posed to the high-risk service, and the display of the representation of the overall risk level includes displaying the list of recommendations.

25. A method for generating a risk map of an optical network, the method comprising: Multiple risks are identified on a computer, representing the risks posed to each link in the optical network due to multiple risk factors, which are determined by simulating the components of the optical network to determine the direct and indirect effects of the multiple risk factors; The remaining margin of the service level is determined based on the risk factors and the current margin of the service, and a confidence interval is determined for each determined remaining margin; A comprehensive risk level representation of the links in the optical network is generated based on the aforementioned multiple risks; as well as This displays a representation of the overall risk level.

26. The method according to claim 25, wherein, Generating a representation of the overall risk level of the links in the optical network includes generating a list of optical network elements that pose a high risk to the network, and wherein displaying the representation of the overall risk level includes displaying the list.

27. The method according to claim 25 or 26, wherein, Generating a representation of the overall risk level of the links in the optical network includes mapping the overall risk level to multiple edges in a graph, wherein each edge in the graph represents the overall risk level of the links in the optical network, and wherein each node in the graph represents an element of the optical network; and wherein displaying the representation of the overall risk level includes displaying the graph.

28. The method according to claim 25 or 26, wherein, Generating a representation of the overall risk level of the links in the optical network includes generating a list of recommendations for reducing the risk of high-risk links, and displaying the representation of the overall risk level includes displaying the list of recommendations.

Images