Service availability determination method, electronic equipment and computer readable medium

Abstract

The invention provides a method for determining service availability, which comprises the following steps of: determining an occurrence event sequence of a state change event according to a preset service model; the preset service model represents a service network structure and a mapping relation between services and service nodes; the state change event comprises at least one of a fault event and a repair event; and determining the service availability of the target service in a preset time period according to the occurrence event sequence. The invention further provides electronic equipment and a computer readable medium.

Description

Technical Field

[0001] The present disclosure relates to the technical field of reliability system engineering, and in particular to a method for determining service availability, an electronic device, and a computer-readable medium. Background Art

[0002] As the industry transforms towards digitalization and intelligence, 5GToB (5G To Business, 5G applications for business) access network has become the preferred technology for enterprises to transform their business due to its low latency, large bandwidth, wide connection and other characteristics. Reliability prediction is to estimate the reliability by mathematical modeling or simulation modeling in the solution demonstration stage, and then use it as the basis for selecting the best solution. In the industry, out of consideration for the economy and security of the business, reliability prediction for network services supported by 5G (5th Generation Mobile Communication Technology, fifth generation mobile communication technology) has become a prerequisite for the successful deployment and operation of 5GToB access network. In the reliability prediction of 5G unmanned driving business, it is necessary to reflect the failure mode of the business and the dynamic behavior under the corresponding networking technology.

[0003] However, the current technology for predicting service reliability only involves a certain device in the system, lacks consideration of the overall networking of 5G unmanned vehicle services and the service assurance technology therein, and lacks consideration of the dynamic change process of service paths. Moreover, the more mature reliability prediction technology is generally used for reliability prediction of mechanical systems and electronic products, and it is difficult to support reliability prediction of 5G unmanned vehicle services. Therefore, there is still a lack of an effective method for reliability prediction of 5G unmanned vehicle services. Summary of the invention

[0004] The embodiments of the present disclosure provide a method for determining service availability, an electronic device, and a computer-readable medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for determining service availability, including:

[0006] According to a preset business model, a sequence of events of state change is determined; the preset business model represents the business network structure and the mapping relationship between business and business nodes; the state change event includes: at least one of a fault event and a repair event; according to the sequence of events, the business availability of the target business within a preset time period is determined.

[0007] In a second aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a memory on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining the service availability described in the first aspect of the embodiment of the present disclosure.

[0008] In a third aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method for determining service availability described in the first aspect of the embodiment of the present disclosure is implemented.

[0009] In the method for determining service availability provided by the embodiment of the present disclosure, a sequence of occurrence events of service status change events is determined by a preset service model, the networking structure of the service is fully considered, time and probability are introduced in the process of determining the service availability of the target service, and the dynamic change process of the service path can be reflected. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a structural diagram of an application environment of a method for determining service availability in an embodiment of the present disclosure;

[0011] Figure 2 is a flow chart of a method for determining service availability in an embodiment of the present disclosure;

[0012] Figure 3 is a flowchart of some steps in another method for determining service availability in an embodiment of the present disclosure;

[0013] Figure 4 is a flowchart of some steps in another method for determining service availability in an embodiment of the present disclosure;

[0014] Figure 5 is a flowchart of some steps in another method for determining service availability in an embodiment of the present disclosure;

[0015] Figure 6 is a simulation schematic diagram of a sequence of events occurring in an embodiment of the present disclosure;

[0016] Figure 7 is a flowchart of some steps in another method for determining service availability in an embodiment of the present disclosure;

[0017] Figure 8 is a flowchart of some steps in another method for determining service availability in an embodiment of the present disclosure;

[0018] Fig. 9 is a schematic diagram of the structure of an electronic device in an embodiment of the present disclosure;

[0019] Fig.10is a structural diagram of an application environment of a method for determining service availability in an example provided by the present disclosure;

[0020] Fig.11 is a schematic diagram of a business network structure in an example provided by the present disclosure;

[0021] Fig.12 This is a schematic diagram of the mapping relationship between the video service and the control service in the service network structure in an example provided by the present disclosure;

[0022] Fig.13 The present invention provides a flowchart of a method for determining service availability in an example. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the method for determining service availability, the electronic device and the computer-readable medium provided by the present disclosure are described in detail below with reference to the accompanying drawings.

[0024] Example embodiments will be described more fully below with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. On the contrary, the purpose of providing these embodiments is to make the present disclosure thorough and complete and to enable those skilled in the art to fully understand the scope of the present disclosure.

[0025] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0026] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0027] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded.

[0028] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless explicitly defined as such herein.

[0029] Here, some application environments of the embodiments of the present disclosure are described:

[0030] The embodiments of the present disclosure can be applied to the business reliability prediction of 5G unmanned driving services and their sub-services, and the business availability can be determined by counting the interruption time of the services. The infrastructure structure and business logic structure corresponding to the business network structure of the preset business model can be adjusted so that the embodiments of the present disclosure can also be applied to ToB (ToBusiness, enterprise-oriented) scenarios under 5G deployment, such as ports, logistics warehousing and other scenarios deployed in 5G.

[0031] In some examples, reference Figure 1 The infrastructure structure of the service can be a 5G-supported unmanned driving service for steel mills. The 5G unmanned driving service corresponds to the 5G network layer access network, namely the SPN (Slicing Packet Network) bearer network. The network components included in the 5G access network are the components of the network, including: PLC (Programmable Logic Controller), camera, CPE (Customer premises equipment) and base station.

[0032] Among them, the PLC module (i.e., logic controller system) and the camera are located on the unmanned overhead crane in the steel plant, which are used to collect, receive, process and control video data; CPE is the wireless terminal access equipment; the base station is an important part of the wireless access network, which consists of four parts: BBU (Building Base band Unit, indoor baseband processing unit), which is used for signal modulation; RRU (Remote Radio Unit, remote radio frequency module), which is used for radio frequency processing; feeder, which is used to connect RRU and antenna; antenna, which is used for conversion between guided waves on cables and space waves in the air.

[0033] Among them, in the BBU, the main control board provides signaling processing and resource management functions for other boards, and realizes the control of each board within the system; the baseband board is used to realize the baseband signal processing function, and can support the baseband processing related functions of the carrier; the NE (Net Element) board is an edge computing node, which is used to provide core network function redundancy.

[0034] Based on the above-mentioned infrastructure structure for the 5G-supported unmanned driving business in steel mills, a preset business model for the unmanned driving business in steel mills can be constructed, and the business availability of the business can be determined based on the preset business model.

[0035] Figure 2 The present invention is a flowchart of a method for determining service availability according to an embodiment of the present invention.

[0036] First, refer to Figure 2 , the present disclosure provides a method for determining service availability, including:

[0037] S1. Determine the occurrence event sequence of the state change event according to a preset business model; the preset business model represents the business network structure and the mapping relationship between the business and the business node; the state change event includes: at least one of a fault event and a repair event;

[0038] S2. Determine the service availability of the target service within a preset time period according to the sequence of events.

[0039] In an embodiment of the present disclosure, a preset business model represents a model of a business network structure and a mapping relationship between each business and a business node, wherein the business network structure represents an infrastructure structure of a business network, facilities in the infrastructure structure are converted into nodes in the business network structure, and the mapping relationship between a business and a business node represents a mapping relationship between a business and a facility.

[0040] The following is an explanation of building a preset business model:

[0041] The modeling object NEO (Network Entity Operation) includes the infrastructure and business logic structure.

[0042] The infrastructure structure can characterize the physical topology of the network and can be represented by an undirected graph G(V,E), where V = {s 1 ,s 2 ,…,s n} represents the node set of the business network structure, E = {e 1 ,e 2 ,…,e m} represents the link set of the business network structure. For node s i or Link e i , you can define its state, MTBF (Mean Time between Failures), MTTR (Mean Time to Repair), and the service set it carries. and capacity c i In one example, node s i The properties are Among them, state i is 1, indicating that the node i is available, state i is inf, indicating that the node i is unavailable.

[0043] The service logic structure represents the different services carried in the service network structure. The attribute information of the service includes: service type, source node and sink node of the service, service path, etc. The service logic structure of a service can be represented by a tuple, that is, APP = (APP type ,APP od ,APP path ,APP level …), where APP type It is an integer used to represent different types of services. The source node and sink node of the service are APP od =(s o ,s d ), the business path is APP path =(s o ,e j ,…,s i ,…,e k ,s d ), APP level Indicates the service level. In some examples, APP level Used to allocate APPs according to service levels when the link capacity is limited and there are multiple services competing for the link capacity. level order to give priority to services with higher service levels.

[0044] According to the infrastructure structure and business logic structure, the preset business model is G(t i )=g(t i )=(G,APP), where G(t i ) and g(t i ) represent the preset business model, G represents the business network structure, and APP represents the business.

[0045] The state change event is NEC (Network Element Correlation), which is a factor driving the dynamic change of the network state, including at least one of a fault event and a repair event. Among them, a fault event refers to an interruption event caused by a fault factor in the service network structure. The fault factor here can be a node (i.e., facility) or link failure caused by changes in the network environment, human factors, etc. It can be used to characterize the change in the state of the modeled object caused by the fault event, which can be expressed as Γ: G(t i )×EC i →G(t i+1 ), that is, the network state G(t i )In the event EC i Driven by, the state changes to G(t i+1 ).

[0046] A repair event is an event generated relative to a fault event. Specifically, a repair event refers to an event that is repaired and restored for a fault event that has occurred. In some examples, a state change event may also include a recovery strategy event, which refers to triggering a recovery strategy to restore business when a fault event has occurred and has not been restored.

[0047] Business availability refers to the proportion of time that a business can operate normally (i.e. without interruption) within a preset time period. The simulation of the time nodes of state change events is realized by the business network structure based on the preset business model through the occurrence of event sequences. The probability of time and state change events is introduced into the occurrence event sequence, so that the dynamic change process of the business path after the facility failure can be reflected, which makes up for the deficiency of reliability calculation only for the network facility level and only for the static topology.

[0048] Accordingly, in some embodiments, reference Figure 3 , S1 includes:

[0049] S11, determining the occurrence time distribution of the state change event in the preset business model;

[0050] S12. Determine an occurrence event sequence of the state change event according to the occurrence time distribution.

[0051] In the embodiments of the present disclosure, a state change event refers to a change in the state of a facility when a service is running in an infrastructure structure. For a fault event, the fault event can be sampled according to the fault distribution of the preset service model to determine the sequence of events. The fault of a facility (component) is a functional fault, which is specifically manifested as a change in the network state or service network structure caused by a fault or maintenance. The change indicates that a node or link fails / repairs at a certain moment, that is, the fault / repair event generated in the preset service model N is Among them, EC i For node s i Failure / repair events, t i Indicates the time when the failure / repair event occurs (i.e. the time when the state changes), and Represents the attributes of an event. Each event includes at least one attribute. The fault / repair condition generated by the preset business model N can also be written as or Represents node s i In t i Failure events occur at all times or repair incident in, Indicates a fault event. In some examples, a failure event or a repair event may also be a link e i Appear.

[0052] Wherein, in some embodiments, reference Figure 4 , S11 includes:

[0053] S111, determining the occurrence probability distribution of the state change event in the preset business model;

[0054] S112: Determine the occurrence time distribution of the state change event in the preset business model according to the occurrence probability distribution.

[0055] The embodiments of the present disclosure do not impose any special restrictions on the occurrence time distribution of state change events. In some examples, the occurrence time distribution is the inverse function of the probability distribution function U=F(X) (i.e., the occurrence probability distribution) obtained by fitting the failure event or repair event on the preset business model, where the threshold of U is [0,1], that is, U is a random variable on [0,1], representing the cumulative failure probability or maintainability (i.e., repair probability) of the facility or network component. That is, the occurrence time distribution of the state change event is the inverse function X=F(X) of U. -1 (U).

[0056] In some embodiments, S112 includes:

[0057] Determine the occurrence time distribution of the fault event in the preset business model according to the average interval time corresponding to the fault event and the occurrence probability distribution;

[0058] And / or, determining the occurrence time distribution of the repair event in the preset business model according to the average maintenance time corresponding to the repair event and the occurrence probability distribution.

[0059] In this embodiment, according to the probability distribution of occurrence of different state change events, the mean time between failures (MTBF) corresponding to failure events and the mean time to repair (MTTR) corresponding to repair events are respectively substituted into the formula to obtain the occurrence time distribution of failure events and repair events in the preset business model.

[0060] According to the distribution of the occurrence time distribution X, sampling X can obtain the corresponding fault event set or repair event set (i.e., occurrence event sequence). In the case where the component can be repaired, the state change can be reflected according to the occurrence event sequence of the fault event and / or repair event.

[0061] Specifically, in some embodiments, referring to Figure 5, when the occurrence time distribution includes a first occurrence time distribution of the fault event and a second occurrence time distribution of the repair event, S12 includes:

[0062] S121. Determine at least one first occurrence time corresponding to the fault event within a preset period according to the first occurrence time distribution;

[0063] S122, determining at least one second occurrence time corresponding to the repair event within a preset period according to the second occurrence time distribution;

[0064] S123: Sort the first occurrence time and the second occurrence time to determine an occurrence event sequence of the state change event within the preset period.

[0065] In this embodiment, the component s corresponding to the node i The failure time follows the distribution U f =F f (t f ), the maintenance time follows the distribution U m =F m (t m ), the initial running time of the component is t=0.

[0066] According to U f The distribution of t generates a random number (i.e., performs a sampling), which represents the time when the fault event occurs. f =F f -1 (U f ) Get the normal working duration of the component If the sum of the initial operation time and the time when the fault event occurs is less than the preset time period, (T is the preset time period), indicating that a fault event occurs within the preset time period, and the state change time corresponding to the fault event is Here It represents the accumulation of the initial running time and the normal working duration, that is, the occurrence time corresponding to the fault event, and also records the state change time t corresponding to the fault event. i = t, the accumulated t value (the state change time t corresponding to the fault event) i ) can be used as the initial running time for the next calculation of the state change moment of the fault event and / or repair event to obtain the conditions that cause the fault On the contrary, if the simulation period is exceeded, sampling ends.

[0067] Since the repair event occurs after the fault event occurs, that is, the repair event needs to appear when the network status of the node or link is faulty, therefore, if the sampling of the fault event does not exceed the simulation period, then according to U m Generate random numbers from the distribution of t m =F m -1 (U m ) Get the network status of the component during the fault duration If the sum of the initial running time and the time when the repair event occurs is less than the preset time period, (t is the state change time of the last sampled fault event i , T is the preset time period), indicating that a repair event occurs within the preset time period (the occurrence of a repair event can restore the component), then the state change time corresponding to the repair event is Here Indicates the state change time t of the last sampled fault event i The accumulation of the fault duration, that is, the occurrence time of the repair event corresponding to the last sampled fault event, and the state change time t corresponding to the repair event are recorded at the same time. i ′=t, the accumulated t value (the state change time t corresponding to the repair event) i ′) can be used as the initial running time for the next calculation of the state change moment of the fault event and / or repair event to obtain the network repair condition On the contrary, if the simulation period is exceeded, sampling ends.

[0068] Furthermore, the aforementioned sampling of fault events and repair events is performed cyclically in sequence until the currently accumulated event t is greater than or equal to T, and the sampling is terminated. In some examples, the preset time period T is the time range for evaluating the service availability, that is, the sampling evaluates the availability of the service within the preset time period T.

[0069] The state change time t of the sampled fault event 1 ,t 2 ,…,t m (i.e. at least one first occurrence moment), and the state change time t′ of the repair event 1 ,t′ 2 ,…,t′ n (i.e., at least one second occurrence time) to obtain a sequence of fault state changes and a sequence of repair state changes at the infrastructure layer of the service, which together constitute a sequence leading to network failure / repair, i.e., T (1) ≤T (2) ≤…≤T (m+n)(i.e., a sequence of events occurs). In some examples, the state change time t 1 ,t 2 ,…,t m and state change time t′ 1 ,t′ 2 ,…,t′ n The sorting is done from small to large.

[0070] In some embodiments, the state change event further includes a recovery strategy event, and the recovery strategy event is triggered at the first occurrence time and / or the second occurrence time;

[0071] In the case where the state change event includes a recovery strategy event, S12 includes:

[0072] Determine at least one third occurrence time for triggering the recovery strategy event according to the triggering probability of the first occurrence time and / or the second occurrence time of the recovery strategy event;

[0073] The first occurrence time, the second occurrence time and the third occurrence time are sequenced to determine an occurrence event sequence of the state change event within the preset period.

[0074] In the embodiments of the present disclosure, the recovery strategy event is a safeguard measure after a business state change occurs. In some embodiments, the recovery strategy may be a business recovery strategy pre-set in the program. The recovery strategy event is an event triggered after a failure event or a repair event occurs.

[0075] There may be one or more recovery strategies, and when triggered, they may be selectively triggered according to a preset priority, or they may be randomly triggered according to the needs of the target business, and the present disclosure is not limited thereto.

[0076] The service recovery strategy is pre-set based on the reliability requirements of the network and node levels to ensure the service after a component failure occurs in the service network structure. Different recovery strategies have different mapping relationships. In some examples, a backup path for the service is pre-set. After a failure occurs that causes the service to be interrupted, the service is directly migrated to the pre-set backup path. G(t i+1 )=Γ 1 (EC,G(t i ),ρ,τ), where the mapping relationship Γ 1 (·) indicates selecting an alternate path path i And make the business connected process, the path i The selection can be based on priority or randomly selected based on meeting business needs.

[0077] It is worth noting that in some examples, not every failure event or repair event will be triggered. The triggering process may be unable to execute due to reliability design defects and other factors, resulting in the inability to be triggered even if a recovery strategy event is pre-set for a failure event or repair event.

[0078] In addition, it should be noted that the difference between recovery strategy events and repair events is that repair events are for repairing faulty nodes or links, and the corresponding effect after repair is node or link recovery, while recovery strategy events are for target services where nodes or links fail during execution. After the recovery strategy is executed, the faulty nodes or links are not repaired, but a business path that can continue to be executed is replaced so that the target service can continue to be executed. The effect of both after execution is that the target service can continue to run, but the processing objects are different.

[0079] After determining the third occurrence time of the event that triggers the recovery strategy, the sequence of events can be determined based on the first occurrence time of the fault event, the second occurrence time of the repair event and the third occurrence time. The sequence of events can be used to input a preset simulation model for simulation processing to determine the interruption time within a preset time period.

[0080] In one example, referring to Figure 6 , Figure 6 The moment corresponding to the component failure event is the first occurrence moment, the component recovery moment of the component failure event is the second occurrence moment, and the moment corresponding to the business recovery is the third occurrence moment. It can be seen that after the component fails, by executing the recovery strategy event, the business can continue to run by switching the business path; after the repair event is executed, the original faulty component is restored, thus shortening the interruption time of the business after the component fails.

[0081] In some embodiments, determining at least one third occurrence time for triggering the recovery strategy event according to the triggering probability of the recovery strategy event being triggered at the first occurrence time and / or the second occurrence time includes:

[0082] Obtaining a business recovery strategy corresponding to the recovery strategy event, wherein the business recovery strategy includes: a state change event triggering the business recovery strategy, a business state, a trigger probability, and a time consumed to trigger the business recovery strategy;

[0083] In the case of determining that the recovery policy event is triggered, the third occurrence time is determined according to the time consumed by triggering the service recovery policy.

[0084] In this embodiment, the recovery strategy event is a state change event Γ: G(t i)×EC i →G(t i+1 ), which can be specifically described as G(t i+1 )=Γ(EC,G(t i ),ρ,τ), represents the current network state G(t i ), if there is a failure event or a repair event EC, then there is a probability ρ that the recovery strategy event is successfully executed, and the event of executing the recovery strategy event is τ. At this time, the network state is G(t i+1 ). Among them, EC represents the failure event or repair event used to trigger the recovery strategy event, G(t i ) is the current (i.e. t i G(t) represents the network state (i.e., business state) at the time, ρ represents the probability of successful execution of the recovery strategy event (i.e., the triggering probability), and τ represents the time consumed by the execution of the recovery strategy event (i.e., the time consumed by triggering the business recovery strategy). In some examples, for a business that adopts a dual-active design, when a component of its business network structure has a failure event, there is a component that is also working and can carry the business corresponding to the failure event. At this time, the time consumed by the execution of the recovery strategy event corresponding to the business recovery strategy event is considered to be 0, that is, τ = 0. Among them, G(t i+1 ) i+1 Indicates the third occurrence time.

[0085] In one example, at t i At time t, a fault event EC of component A occurs, and the corresponding node in the business network structure fails. At this time, the network state is G(t i ).

[0086] When there is a cold backup component A' in component A, the path of the target service is switched from the original path to the new path where the cold backup component A' is located, which is a recovery strategy event, and its mapping relationship is Γ(·). The path switching success rate of the target service is ρ, and the switching time is τ. After the switching is completed (that is, after the recovery strategy event is successfully triggered), the network state is updated to G(t i+1 ).

[0087] By introducing recovery strategy events, the business networking mode, specific failure mode of the business, sub-business assurance technology and recovery strategy are fully considered, avoiding over-simplification of the model, so that the business characteristics can be fully reflected, and the calculation results of business availability are closer to the actual situation.

[0088] In some embodiments, reference Figure 7 , S2 includes:

[0089] S21, determining the interruption time of the target service according to the occurrence event sequence of the state change event; the interruption time includes multiple time intervals during which the target service has a failure event and has not been repaired;

[0090] S22: Determine the service availability of the target service within the preset time period according to the interruption time.

[0091] Reference Figure 6 In the embodiments of the present disclosure, the interruption time of a service refers to a plurality of time intervals during which a fault event occurs in a target service and is not repaired. The occurrence of a fault event and not repaired may be a first time period between the occurrence of the fault event and the execution of a corresponding recovery strategy event, or a second time period between the occurrence of the fault event and the execution of a repair event. In some examples, each time interval in the interruption time corresponds to each occurrence of a fault event. If a recovery strategy event is triggered when a fault event occurs, the time interval of the interruption time corresponding to the fault event is the first time period. If a recovery strategy event is not triggered when a fault event occurs, the time interval of the interruption time corresponding to the fault event is the second time period.

[0092] Business availability refers to the proportion of time that the target business can operate normally within a preset time period. When the interruption time is determined, the preset time period is the normal operation time except for the interruption time caused by the fault event. Therefore, after determining the interruption time, the business availability can be determined.

[0093] In some embodiments, reference Figure 8 , S22 includes:

[0094] S221, obtaining the service availability of the target service for multiple occurrences of the event sequence within a preset time period;

[0095] S222: Taking the average value of the service availability as the target service availability of the target service in the preset time period.

[0096] In this embodiment, the occurrence event sequence is obtained by sampling based on the occurrence time distribution of the state change event. Each occurrence event sequence obtained can be used as the operation simulation of the target business within a preset time period. Therefore, when multiple samplings are performed to determine the occurrence event sequence, the results of multiple simulations (i.e., the business availability of multiple occurrence event sequences) are averaged as the target business availability. The more occurrence event sequences are determined by simulation, the higher the accuracy of the business availability.

[0097] In some embodiments, when the target service includes multiple sub-services, determining the service availability of the target service within the preset time period according to the interruption time includes:

[0098] An average value of the service availability of the sub-services in the preset time period is used as the service availability of the target service in the preset time period.

[0099] In the embodiment of the present disclosure, for the service availability of a target service composed of multiple sub-services, the average of the service availabilities of all sub-services is taken.

[0100] It is worth noting that the embodiments of the present disclosure can also perform reliability evaluation on different services under the same networking solution to comprehensively consider the impact of failures of software and hardware facilities in the service network structure on service availability.

[0101] In one example, multiple occurrence event sequences are input into a preset simulation network model, and the preset simulation network model is used to simulate the target business corresponding to the occurrence event sequence and the state change time of each occurrence within a preset time period, and trigger the business recovery strategy event at the state change moment corresponding to each fault event or repair event to drive the business state change, record the duration of each business state, and count the business interruption time;

[0102] pass Calculate the service availability corresponding to the N-times event sequence, where: is the service availability of the i-th target service, N is the total number of times, For the interruption time, It is the normal operation time (i.e. business connectivity time).

[0103] In another example, by Calculate the service availability of the target service composed of multiple sub-services, where A con is the service availability of the target service, N app is the total number of sub-businesses, is the service availability of the i-th sub-service.

[0104] In the method for determining the service availability corresponding to the embodiment of the present disclosure, by inputting the mean time between failures MTBF, the mean time to repair MTTR, the infrastructure structure topology of the service networking, the logical structure of the service and the recovery strategy (redundant backup strategy) of each network component, the network composition, service failure mode and service assurance technology (i.e., recovery strategy) are fully considered, and the failure or recovery state of the network state of the target service in the service network structure that evolves over time is dynamically modeled, thereby estimating the service availability. The availability can characterize the operational reliability of the service on the service network structure, and provide a theoretical basis for selecting a suitable networking solution for the service.

[0105] Thirdly, refer to Fig. 9, an embodiment of the present disclosure provides an electronic device, comprising:

[0106] One or more processors 901;

[0107] A memory 902 having one or more programs stored thereon, which, when executed by one or more processors, enables the one or more processors to implement any one of the above methods for determining service availability;

[0108] One or more I / O interfaces 903 are connected between the processor and the memory and are configured to implement information exchange between the processor and the memory.

[0109] Among them, the processor 901 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 902 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) 903 is connected between the processor 901 and the memory 902, and can realize information interaction between the processor 901 and the memory 902, including but not limited to a data bus (Bus), etc.

[0110] In some embodiments, the processor 901 , the memory 902 , and the I / O interface 903 are connected to each other via a bus 904 , and further connected to other components of the computing device.

[0111] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the determination of any of the above-mentioned service availability is implemented.

[0112] In order to enable those skilled in the art to more clearly understand the technical solutions provided by the embodiments of the present disclosure, the technical solutions provided by the embodiments of the present disclosure are described in detail below through specific embodiments:

[0113] In one example, for the 5G unmanned driving service, a model is built based on the infrastructure structure topology of the service networking related to the service, the logical structure of the service, the conditions for service failure events, and the recovery strategy to obtain a preset service model, and reliability determination is implemented based on the preset service model. More specifically, the process of determining service availability is as follows:

[0114] The infrastructure structure and business logic structure of the 5G unmanned driving service are modeled, that is, the modeling object NEO includes the infrastructure and business logic structure of the 5G driving service.

[0115] The infrastructure structure represents the physical topology of the network and can be represented by an undirected graph G(V,E), where V = {s 1 ,s 2 ,…,s n} represents the node set of the network, E={e 1 ,e 2 ,…,e n} represents the link set of the network. For node s i or Link e i , you can define its state, MTBF, MTTR, and the service set it carries and capacity c i Related attributes, such as Among them, state is 1 for available, and state is inf for unavailable.

[0116] The service logic structure represents different services carried in the network. The service attribute information includes service type, service source node and sink node, service path, and service APP=(APP type ,APP od ,APP path ,APP level …), where APP type Used to represent different types of business, APP type The value of APP is an integer. od =(s o ,s d ), APP od Indicates the source node s of the service o and sink nodes d , APP path =(s o ,e j ,…,s i ,…,e k ,s d ), APP path Indicates the service path, e j and e k All are links, s i is node i.

[0117] According to the infrastructure structure and business logic structure modeling, the network state (i.e., the preset business model) is G(t i )=g(t i )=(G,APP).

[0118] According to the preset business model, the sequence of occurrence events of fault events and repair events is determined. The fault events and repair events here are factors that drive the change of network status in the 5G unmanned driving business. The fault event-driven network status refers to the node failure and / or link failure caused by the change of the network environment and human factors, which in turn causes the change of status. It can be mapped by Γ:G(t i )×EC i →G(t i+1 ) indicates that the mapping represents the network state G(t i )In the event EC i Driven by, the network state changes to G(t i+1 ).

[0119] The occurrence time distribution of the state changes of each component (i.e., facility) in the infrastructure structure with respect to the failure event is sampled:

[0120] The failure event of a component in the network is a functional failure, which is specifically manifested as a failure event or repair event of a network component causing a change in the network state, that is, a node or link fails or is repaired at a certain time. The conditions for the occurrence of a failure event or repair event in model N are Can write fault events or repair incident Represents node s i (or link e i ) at t i Failure events occur at all times or repair incident

[0121] The corresponding probability distribution function F(X) is obtained by fitting the data of component failure events or repair events. Set U as a random variable with a threshold on [0,1], and U represents the cumulative failure probability or repair degree of the component. Let U = F(X), and the probability distribution function of U can be obtained based on F(X). The inverse function of F(X) is calculated as X = F -1 (U) According to the occurrence time distribution obeyed by X, X is sampled to obtain the corresponding fault event set or repair event set; further, if the component is repairable, the state change of the network component can be determined according to the fault event set or the repair event set.

[0122] More specifically, let component s i The fault events follow the occurrence time distribution U f =F f (t f ), the repair event follows the occurrence time distribution U m =F m (t m), the initial working time of the component is t = 0. According to the time distribution of the fault event, The normal working duration of the component is obtained as

[0123] like (T is the preset time period), then the current accumulated time The accumulated time of the current fault event occurrence is taken as the state change time t i And record, the conditions that lead to the fault event are On the contrary, if the preset time period is exceeded (i.e. ), the component status sampling ends.

[0124] If the preset time period is not exceeded (i.e. ), then according to the occurrence time distribution of the repair event, Get the duration of the fault state after the component failure event like but The accumulated time of the current repair event occurrence is taken as the state change time t i ′ and record it, and the generation condition of the repair event after the fault event is obtained as follows:

[0125] Furthermore, the aforementioned sampling operation of the fault events and the repair events is repeated cyclically until the accumulated time of the sampled events exceeds a preset time period.

[0126] The state change of the sampled fault event corresponds to the occurrence event sequence t 1 ,t 2 ,…,t m , and the occurrence event sequence t′ corresponding to the state change of the repair event 1 ,t′ 2 ,…,t′ n Arranged in descending order, the two together constitute the sequence of events that cause the network state to change: T (1) ≤T (2) ≤…≤T (m+n) .

[0127] According to the pre-set service recovery strategy, a recovery strategy event is triggered after a failure event or a repair event occurs. The service recovery strategy is a description of the further service status change process after the status of the 5G unmanned driving service changes. When a network failure event or a repair event occurs, the corresponding recovery strategy event will be triggered accordingly. In addition, the recovery strategy event may not be triggered due to reliability design defects, resulting in the service being unable to recover from the failure state. Based on this, the service recovery strategy of the 5G unmanned driving service is Γ:G(ti )×EC i →G(t i+1 ), can also be expressed as: G(t i+1 )=Γ(EC,G(t i ),ρ,τ), indicating that the current network state of the service is G(t i ), if there is a network component failure event EC in this state, there is a probability ρ of executing the recovery strategy event. The time for executing the recovery strategy is τ, and the state of the business after the execution is G(t i+1 ). Among them, EC represents a failure event or a repair event, G(t i ) is the network status of the service, ρ represents the probability of the recovery strategy event being triggered, and τ represents the time consumed in the execution process of the recovery strategy event.

[0128] If the business is in t i At time t, a component failure event EC occurs, and the corresponding node in the network fails. At this time, the network state is G(t i ). If the component has a cold backup component, the service path is switched from the original path to the new path where the cold backup component is located, that is, there is a recovery strategy, which can be expressed as a mapping relationship Γ(·), the success rate of service switching is ρ, and the switching time is τ. After the switching is completed, the network state is updated to G(t i+1 ).

[0129] The recovery strategy pre-set for the service can be set as different mapping relationships Γ(·) from the perspective of the reliability of the network level and the component level after the component failure event. More specifically, before the component fails, the backup path for the service is pre-set. When a failure event occurs and the service is interrupted, the service is migrated to the backup path. This process can be expressed as: G(t i+1 )=Γ 1 (EC,G(t i ),ρ,τ), where the mapping Γ 1 (·) is to select a backup path i , used to make services connected after a failure occurs, for path i The selection can be based on priority or randomly selected based on meeting business requirements.

[0130] For a service that adopts a dual-active design, when a component function fails, if there is a component that is also working and can carry the service on the failed component, the execution time of the recovery strategy event of the service can be considered to be 0, that is, τ=0.

[0131] See also Figure 6, the mapping relationship between the occurrence event sequence corresponding to the fault event and the repair event is input into the preset simulation model, and at the moment of each state change, it is determined whether to trigger the recovery strategy event, drive the unmanned driving business state change, and record the duration of each business state, and count the business connectivity interruption duration (i.e., business terminal duration).

[0132] Repeat the above process of determining the sequence of events, and calculate the duration of the target service's connectivity interruption N times. and duration of connectivity interruption Determine the connectivity availability of the target service during N simulations as

[0133] If the target service consists of multiple sub-services, the average connectivity availability of the sub-services is taken. Connectivity availability as the target service.

[0134] The above process determines the occurrence event sequence of the service state change event through a preset service model, fully considers the network structure of the service, introduces time and probability in the process of determining the service availability of the target service, and can reflect the dynamic change process of the service path.

[0135] Reference Fig.10 In another example, the business infrastructure structure corresponding to the 5G unmanned driving business of a steel plant is as follows: Fig.10 As shown, the service infrastructure structure is a dual-frequency networking structure, that is, in the event of a failure in a 5G module for 3.5G transmission or a 5G module for 2.1G transmission, transmission can be performed through another 5G module.

[0136] According to the business infrastructure structure, the MTBF and MTTR of network components such as CPE common parts, 5G modules, RRUs, fronthaul links, baseband boards, main control boards, power boards, edge nodes (NE boards) and backhaul links are determined; at the same time, the business network structure G (V, E) corresponding to the business infrastructure structure is constructed, and the four business logic structures are determined as APP 1 、APP 2 、APP 3 And APP 4 .

[0137] Reference Fig.10 , Fig.11 as well as Fig.12 Specifically, Fig.10Each device in the dual-frequency networking infrastructure structure is converted into a node, an undirected graph with 28 nodes is established, and the service network structure corresponding to the unmanned vehicle is modeled. The service initial path is determined according to the service logic structure, and the service-to-service node mapping relationship of the service network structure corresponding to the unmanned vehicle is modeled. According to the service network structure and the service-to-service node mapping relationship, the preset service model is determined. Among them, the service includes video service and control service.

[0138] The process of determining the service availability of the unmanned vehicle service is as follows: Fig.13 As shown, refer to Fig.13 , simulate the 5G unmanned driving service N times, initialize the number of simulations, set the number of simulations i = 1, increase it successively, set the confidence level to 95%, set the simulation time T to 10 years, and set the error threshold to 1e-4. The maximum number of simulations is 300.

[0139] Input the initial state of the 5G driving service infrastructure structure and business logic structure (preset business model), determine the occurrence event sequence of the state change event corresponding to the preset business model, and the state change event includes the failure event and the repair event, namely EC V ={EC V,i},i∈N, In the case of k = F, represents a fault event. In the case of k = M, Indicates a repair event.

[0140] The failure event and the repair event both obey the exponential distribution, that is, Solving the inverse function for U yields Among them, U follows a uniform distribution in [0,1].

[0141] Determine the 1 / MTBF corresponding to the failure event and the 1 / MTTR corresponding to the repair event, and use them as input The occurrence time distribution of the fault event and the repair event is obtained by λ in t. The occurrence time distribution is sampled to obtain the occurrence event sequence of the dynamic simulation of all components. i The occurrence event sequence of the state change events is obtained by sorting them in descending order. That is, the occurrence event sequence includes multiple samplings of the occurrence time of the fault events and / or the repair events, and the occurrence time of the sampled fault events and / or the repair events are arranged according to a preset rule to obtain the occurrence event sequence.

[0142] The sequence of events is input into the simulation model. Specifically, the simulation model extracts the current event from the sequence of events to update the network topology of the business:

[0143] Case 1: When the event is a fault event, the set of faulty network components is updated, and it is determined whether the component service based on the fault is interrupted; in the case of service interruption, it is further determined whether there are redundant resources at the fault location of the service interruption for executing the recovery strategy event; in the case that there are redundant resources at the fault location of the service interruption for executing the recovery strategy event, the service is further restored according to the recovery strategy event, and the service interruption time is updated from the time period between the fault event and the completion of the recovery strategy event, and the network topology of the service (i.e., the service network structure) is updated at the same time;

[0144] For case 1, a service recovery strategy is pre-set in the program, including the probability and time of service migration when each node fails. The recovery strategy includes: CPE1+1, RRU1+1, fronthaul link 1+1, baseband board 1+1, main control board 1+1, power board 1+1, backhaul link 1+1, UPF+ (Enhanced User Plane Function+). CPE1+1, fronthaul link 1+1, power board 1+1, backhaul link 1+1 and UPF+ are all active-active. Active-active means that service recovery can be performed without delay / execution time, that is, τ=0; baseband board 1+1 and main control board 1+1 are both cold standby designs, which are recovery strategies based on active protection, that is, G(t i+1 )=Γ 1 (EC,G(t i ),ρ,τ), where ρ and τ are set according to the service migration probability and service migration time, see Table 1:

[0145] Table 1

[0146]

[0147] Case 2: if the event is a fault event, update the set of faulty network components and determine whether the service of the component based on the fault is interrupted; if the service is not interrupted, update the network topology of the service;

[0148] Case 3: When the event is a fault event, the set of faulty network components is updated, and it is determined whether the component service based on the fault is interrupted; in the case of service interruption, it is further determined whether there are redundant resources at the fault location of the service interruption for executing the recovery strategy event; in the case that there are no redundant resources at the fault location of the service interruption for executing the recovery strategy event, the network topology of the service is updated;

[0149] Case 4: when the event is not a fault event, the faulty network component set is updated, the restored components are added to the network topology G of the service, and the network topology of the service is updated;

[0150] Repeatedly take out the current event from the occurrence event sequence until the occurrence event sequence is empty, thus completing the simulation of the occurrence event sequence.

[0151] The simulation of the event sequence is repeated N times, and the service availability is calculated based on the N simulations, and the average value of the N service availabilities is used as the target service availability of the unmanned vehicle service.

[0152] The service availability is calculated by counting the total time of 5G unmanned driving service interruption. The target service availability results are shown in Table 2:

[0153] Table 2

[0154]

[0155]

[0156] It can be seen that the method for determining the service availability of 5G unmanned vehicle services establishes a model of the network infrastructure structure and the mapping relationship between services and facilities, and uses the failure events and / or repair events obtained based on equipment failure and repair data sampling as the input of the simulation model to drive the state changes of network infrastructure and services, thereby realizing the simulation of dynamic business behavior. The service availability is also obtained by statistics on the service interruption time, that is, the proportion of time that the service can operate normally. This can solve the problem of reliability prediction of 5G unmanned vehicle services, provide a more effective reliability evaluation for the networking solution of the service, guide the further improvement of the solution, and at the same time, provide customers with more reliable business networking solutions that comply with SLA (Service Level Agreement).

[0157] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0158] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly noted, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. A method for determining service availability, comprising: According to the preset business model, determine the occurrence sequence of state change events; The preset business model represents the business network structure and the mapping relationship between business and business nodes; The state change event includes: at least one of a failure event and a repair event; The service availability of the target service within a preset time period is determined according to the sequence of events that occur.

2. The method for determining service availability according to claim 1, wherein: According to the preset business model, determine the sequence of events for state change events, including: Determining the occurrence time distribution of the state change event in the preset business model; According to the occurrence time distribution, an occurrence event sequence of the state change event is determined.

3. The method for determining service availability according to claim 2, wherein: Determining the occurrence time distribution of the state change event in the preset business model includes: Determine the occurrence probability distribution of the state change event in the preset business model; According to the occurrence probability distribution, the occurrence time distribution of the state change event in the preset business model is determined.

4. The method for determining service availability according to claim 3, wherein: Determining the occurrence time distribution of the state change event in the preset business model according to the occurrence probability distribution includes: Determine the occurrence time distribution of the fault event in the preset business model according to the average interval time corresponding to the fault event and the occurrence probability distribution; And / or, determining the occurrence time distribution of the repair event in the preset business model according to the average maintenance time corresponding to the repair event and the occurrence probability distribution.

5. The method for determining service availability according to claim 2, wherein: In a case where the occurrence time distribution includes a first occurrence time distribution of the fault event and a second occurrence time distribution of the repair event, determining an occurrence event sequence of the state change event according to the occurrence time distribution includes: Determining at least one first occurrence time corresponding to the fault event within a preset period according to the first occurrence time distribution; Determining at least one second occurrence time corresponding to the repair event within a preset period according to the second occurrence time distribution; The first occurrence time and the second occurrence time are sequenced to determine an occurrence event sequence of the state change event within the preset period.

6. The method for determining service availability according to claim 5, wherein: The state change event also includes a recovery strategy event, and the recovery strategy event is triggered at the first occurrence time and / or the second occurrence time; In the case where the state change event includes a recovery strategy event, determining the occurrence event sequence of the state change event according to the occurrence time distribution includes: Determine at least one third occurrence time for triggering the recovery strategy event according to the triggering probability of the recovery strategy event being triggered at the first occurrence time and / or the second occurrence time; The first occurrence time, the second occurrence time and the third occurrence time are sorted to determine an occurrence event sequence of the state change event within the preset period.

7. The method for determining service availability according to claim 6, wherein: Determining at least one third occurrence time for triggering the recovery strategy event according to the triggering probability of the recovery strategy event being triggered at the first occurrence time and / or the second occurrence time includes: Obtaining a business recovery strategy corresponding to the recovery strategy event, wherein the business recovery strategy includes: a state change event triggering the business recovery strategy, a business state, a trigger probability, and a time consumed to trigger the business recovery strategy; In the case of determining that the recovery policy event is triggered, the third occurrence time is determined according to the time consumed by triggering the service recovery policy.

8. The method for determining service availability according to claim 1, wherein: Determining the service availability of the target service within a preset time period according to the sequence of events, including: Determine the interruption time of the target service according to the occurrence event sequence of the state change event; the interruption time includes multiple time intervals during which the target service has a failure event and has not been repaired; The service availability of the target service within the preset time period is determined according to the interruption time.

9. The method for determining service availability according to claim 8, wherein: Determining the service availability of the target service within the preset time period according to the interruption time includes: Obtaining the service availability of the target service for multiple occurrences of the event sequence within a preset time period; The average value of the service availability is used as the target service availability of the target service in the preset time period.

10. The method for determining service availability according to claim 8, wherein: In the case where the target service includes multiple sub-services, determining the service availability of the target service within the preset time period according to the interruption time includes: An average value of the service availability of the sub-services in the preset time period is used as the service availability of the target service in the preset time period.

11. An electronic device, comprising: one or more processors; A memory having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining service availability according to any one of claims 1 to 10.

12. A computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method for determining service availability according to any one of claims 1 to 10 is implemented.

Images