An alarm management system for content distribution network

Abstract

The present invention belongs to the field of content distribution network, and discloses an alarm management system for content distribution network, including a data collection module, a data analysis module and an alarm module; the data collection module is used to collect the operation parameters of the content distribution network; the data analysis module includes a calculation unit, an analysis unit and an update unit; the calculation unit is used to calculate the priority processing value of the operation parameter based on the correlation value between the operation parameter and the alarm event; the analysis unit is used to analyze the operation parameter based on the priority coefficient, and the update unit is used to update the correlation value between the operation parameter and the alarm event; the alarm module is used to alarm according to a pre-set alarm strategy when an alarm event of a preset type occurs. The present invention can preferentially analyze and calculate the operation parameters with high priority processing values, thereby improving the timeliness of discovering the alarm event.

Description

Technical Field

[0001] The present invention relates to the field of content distribution networks, and in particular to an alarm management system for content distribution networks. Background Art

[0002] Content Delivery Network (CDN) is a distributed network system that aims to deliver content (such as web pages, images, videos, scripts, etc.) to end users more quickly and reliably by deploying server nodes in multiple geographical locations.

[0003] The core idea of ​​CDN is to copy content from the original server to multiple widely distributed cache servers. When a user requests a certain content, CDN will redirect the request to the cache server closest to the user, thereby reducing data transmission delays and bandwidth consumption, and improving user access speed and experience.

[0004] During the operation of CDN, it is usually necessary to monitor it to ensure service quality and performance. When an event that meets the alarm rules is detected, the alarm system will issue a corresponding alarm. In the prior art, when performing alarm management on CDN, one type of alarm event usually contains at least one type of operating parameter, and one type of operating parameter corresponds to a fixed alarm processing priority value for a long time. When multiple types of operating parameters are received at the same time, the higher the priority value, the higher the priority of analyzing the operating parameters of this type. However, since the usage of the CDN network is always changing, the fixed processing priority value can easily lead to alarm events that are not discovered in time for processing, thereby affecting the service quality of the CDN network and the user experience. Summary of the invention

[0005] The purpose of the present invention is to disclose an alarm management system for a content distribution network, so as to solve the problem of how to improve the timeliness of discovering alarm events during the process of alarm management for the content distribution network.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides an alarm management system for a content distribution network, comprising a data collection module, a data analysis module and an alarm module;

[0008] The data collection module is used to collect the operating parameters of the content distribution network;

[0009] The data analysis module includes a calculation unit, an analysis unit and an update unit;

[0010] The calculation unit is used to calculate the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event;

[0011] The analysis unit is used to analyze the operating parameters based on the priority coefficients, including:

[0012] If the number of types of operating parameters received within the set analysis period is greater than or equal to 2, the operating parameters are analyzed in descending order of priority values ​​of the operating parameters to determine whether an alarm event of a preset type occurs;

[0013] If only one type of operating parameter is received within the set analysis period, the operating parameter is analyzed to determine whether an alarm event of the preset type occurs;

[0014] The updating unit is used to update the correlation value between the operating parameter and the alarm event;

[0015] The alarm module is used to generate an alarm according to a preset alarm strategy when an alarm event of a preset type occurs.

[0016] Preferably, collecting the operating parameters of the content distribution network includes:

[0017] The operating parameters of each CDN node in the content distribution network are collected separately.

[0018] Preferably, calculating the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event includes:

[0019] A preset calculation cycle is used to calculate the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event.

[0020] Preferably, the calculation process of the priority value includes:

[0021] Use b to indicate the type of running parameters, and use assval b Indicates the maximum value of the association value between the operating parameter of type b and various types of preset alarm events;

[0022] The calculation formula for the priority processing value of type b operating parameters is:

[0023]

[0024] pripro b represents the priority value of the operating parameter of type b, α1 and α2 represent the preset first weight and second weight respectively, cyc b Indicates that in the time period [coltim b -H,coltim b], the set of b-type operating parameters collected by the data acquisition module, coltim b Indicates the end time of the previous calculation cycle, H is the preset calculation cycle, and i represents cyc b The value of the element in, numi represents cyc b The number of elements in mai represents the maximum value of the operating parameter of type b.

[0025] Preferably, analyzing the operating parameters to determine whether a preset type of alarm event occurs includes:

[0026] The operating parameters are calculated with the alarm rules of various types of preset alarm events input to determine whether the number of alarm rules that the operating parameters meet is greater than or equal to 1. If so, it indicates that an alarm event of the preset type has occurred. If not, it indicates that an alarm event of the preset type has not occurred.

[0027] Preferably, updating the association value between the operating parameter and the alarm event includes:

[0028] An adaptive update cycle is used to update the correlation value between the operating parameters and the alarm events.

[0029] Preferably, the calculation process of the adaptive update period includes:

[0030] Using cycupd b,c+1 Indicates the c+1th update cycle corresponding to the running parameters of type b. After the cth update cycle ends, if numtrig c Greater than numtrig c-1 , then cycupd b,c+1 The calculation formula is:

[0031]

[0032] If numtrig c Less than or equal to numtrig c-1 , then cycupd b,c+1 The calculation formula is:

[0033]

[0034] cycupd b,c Indicates the cth update cycle corresponding to the running parameters of type b, numtrig c and numtrig c-1 Indicates the number of alarm events triggered by the b-type operating parameters in the c-th and c-1-th update cycles, respectively. cIt indicates the total number of b-type operating parameters obtained by the data acquisition module in the c-th update cycle, basic is the basic change duration of the update cycle, cycupdmi and cycupdma respectively indicate the minimum and maximum values ​​of the preset update cycle.

[0035] Preferably, updating the association value between the operating parameter and the alarm event includes:

[0036] For an alarm event of type d, the type of the benchmark operating parameter corresponding to the alarm event of type d is represented as type d ;

[0037] After the cth update cycle is over, use U b,c and Indicates the operating parameters and type of type b obtained in the cth update cycle d A collection of operating parameters of the type;

[0038] Will U b,c and The correlation coefficient between them is used as the association value between the operating parameters of type b and the alarm events of type d.

[0039] Preferably, the alarm module includes an alarm level acquisition unit and an alarm unit;

[0040] The alarm level acquisition unit is used to acquire the alarm level corresponding to the alarm event of the preset type that occurs;

[0041] The alarm unit is used to issue an alarm based on the alarm strategy corresponding to the alarm level.

[0042] Preferably, the alarm strategy includes an alarm method and an alarm content.

[0043] Beneficial effects:

[0044] Compared with the existing content distribution network alarm management method, the present invention updates the correlation value between the operating parameters and the alarm events at irregular intervals, and then calculates the priority processing value regularly. In this way, when the data analysis module receives multiple types of operating parameters within the set analysis cycle, it can give priority to analyzing and calculating the operating parameters with high priority processing values, thereby improving the timeliness of discovering alarm events. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0046] Figure 1 This is a first schematic diagram of an alarm management system for a content distribution network according to the present invention.

[0047] Figure 2 This is a second schematic diagram of an alarm management system for a content distribution network according to the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] like Figure 1 In one embodiment shown, the present invention provides an alarm management system for a content distribution network, including a data collection module, a data analysis module and an alarm module;

[0050] The data collection module is used to collect the operating parameters of the content distribution network;

[0051] The data analysis module includes a calculation unit, an analysis unit and an update unit;

[0052] The calculation unit is used to calculate the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event;

[0053] The analysis unit is used to analyze the operating parameters based on the priority coefficients, including:

[0054] If the number of types of operating parameters received within the set analysis period is greater than or equal to 2, the operating parameters are analyzed in descending order of priority values ​​of the operating parameters to determine whether an alarm event of a preset type occurs;

[0055] If only one type of operating parameter is received within the set analysis period, the operating parameter is analyzed to determine whether an alarm event of the preset type occurs;

[0056] The updating unit is used to update the correlation value between the operating parameter and the alarm event;

[0057] The alarm module is used to generate an alarm according to a preset alarm strategy when an alarm event of a preset type occurs.

[0058] The scheme of the present invention is very different from the prior art. First, the priority processing value is not a fixed value, which makes the priority processing value change with the change of the operating conditions in the content distribution network, so that the priority processing value can more accurately represent the priority processing degree of different types of operating parameters. Secondly, since the acquisition frequency of different types of operating parameters is different, multiple types of operating parameters may be received at the same time in the same analysis cycle. The present invention can calculate the different types of operating parameters received in the same analysis cycle according to the priority processing value, and determine whether a preset type of alarm time occurs, thereby improving the probability of timely discovering abnormal situations in the content distribution network. Finally, the associated value used by the present invention to calculate the priority processing value will be updated, which makes the calculated priority processing value more in line with the latest operating conditions of the content distribution network, providing effective reference data for timely discovering abnormal situations in the content distribution network.

[0059] Preferably, the analysis period is set to 1 minute.

[0060] Specifically, the value of the analysis period can be set according to the communication busyness of the content distribution network. The busier the communication, the smaller the value can be set, which is conducive to timely discovering abnormal situations in the content distribution network.

[0061] In the present invention, the analysis cycles are continuous, that is, after the previous analysis cycle ends, the next analysis cycle begins immediately.

[0062] Preferably, collecting the operating parameters of the content distribution network includes:

[0063] For the latest collected operating parameter z, the correctness verification coefficient of the operating parameter z is calculated using the following formula:

[0064]

[0065] corveri z represents the correctness check coefficient of z, and Ub represents the time period [t z -δ,t z ], the set of running parameters of type b collected, δ represents the set correctness check duration, value z and value h Represents the values ​​of z and h respectively, h is the operating parameter in Ub, value mx Indicates the maximum value of the operating parameters in Ub, tim z,h Indicates the time interval between the acquisition times of z and h, tim z,max The maximum value of the time interval between the collection time of the operating parameters in Ub and the collection time of z;

[0066] Calculate the abnormality value of z:

[0067]

[0068] unnrl z Indicates the abnormality value of z;

[0069] If the abnormality value is greater than a preset abnormality threshold, z is not sent to the data analysis module; otherwise, z is sent to the data analysis module.

[0070] In the process of obtaining the operating parameters, the present invention does not directly save all the operating parameters. Because there may be interference events in the content distribution network that cause abnormal values ​​of network parameters, and the impact caused by the interference events is not continuous, if the alarm is triggered based on such an event type, it is easy to increase the workload. Therefore, the present invention first calculates the correctness check coefficient of z, and then calculates the abnormality degree value based on the correctness check coefficient. When the abnormality degree value exceeds the threshold value corresponding to the universe, it means that the operating parameter of this type is a change caused by the interference event, which belongs to the value under abnormal state. The present invention does not send z to the data analysis module for analysis, thereby effectively reducing the probability of false triggering of alarms.

[0071] When calculating the correctness check coefficient, the smaller the numerical difference between z and h and the shorter the time interval between the acquisition moments, the greater the contribution of h to the calculation result of the correctness check coefficient, thereby increasing the difference in the contribution of different operating parameters to the calculation result and improving the effectiveness of the calculated correctness check coefficient.

[0072] Preferably, the preset abnormality threshold is 0.05.

[0073] Preferably, the operating parameters include the CDN node's used bandwidth, total traffic, back-to-source traffic, number of requests received, etc.

[0074] Operation parameters are values ​​related to the status of CDN nodes that change during the operation of the CDN nodes.

[0075] Preferably, collecting the operating parameters of the content distribution network includes:

[0076] The operating parameters of each CDN node in the content distribution network are collected separately.

[0077] Specifically, in a content distribution network, there are usually multiple CDN nodes. CDN nodes are usually deployed in data centers around the world, and these nodes can be physical servers, virtual servers, or cloud servers. The main function of a CDN node is to forward user requests to the node closest to the user by caching content and using global load balancing technology, thereby speeding up content delivery and improving user experience. In short, a CDN node acts as a cache server, storing a wide range of content copies, allowing users to access the content more quickly rather than getting it from the original server.

[0078] Preferably, calculating the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event includes:

[0079] A preset calculation cycle is used to calculate the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event.

[0080] In the present invention, the preset calculation cycle is continuous, and when the current calculation cycle ends, the next calculation cycle will be entered immediately. After the current calculation cycle ends, the priority processing value is updated based on the operating parameters of the content distribution network of the previous calculation cycle.

[0081] By periodically updating the priority processing value, the priority processing value can be more closely adapted to the latest operation status of the content distribution network.

[0082] Preferably, the preset calculation period is 30 minutes.

[0083] Preferably, the calculation process of the priority value includes:

[0084] Use b to indicate the type of running parameters, and use assval b Indicates the maximum value of the association value between the operating parameter of type b and various types of preset alarm events;

[0085] The calculation formula for the priority processing value of type b operating parameters is:

[0086]

[0087] pripro b represents the priority value of the operating parameter of type b, α1 and α2 represent the preset first weight and second weight respectively, cyc b Indicates that in the time period [coltim b -H,coltim b ], the set of b-type operating parameters collected by the data acquisition module, coltim b Indicates the end time of the previous calculation cycle, H is the preset calculation cycle, and i represents cycb The value of the element in, numi represents cyc b The number of elements in mai represents the maximum value of the operating parameter of type b.

[0088] Since each type of operating parameter may not be related to only one type of alarm event, the maximum value of the associated value of the present invention indicates the importance of the current type of operating parameter. The larger the maximum value of the associated value, the more important the current type of operating parameter is in the process of judging whether an alarm event occurs.

[0089] However, since the state of the content distribution network is always changing, and since the calculation process of the associated value is relatively complicated, the update cycle of the associated value is relatively long. If an update cycle is set too small, the associated value will be frequently calculated, which will occupy more computing resources in the alarm management system, thus affecting the monitoring performance of the alarm event. Therefore, the priority processing value of the present invention is calculated based on the associated value and combined with the operating parameters in the previous calculation cycle. When the fluctuation of the operating parameters in the previous calculation cycle is greater, the priority processing value will be correspondingly larger, and vice versa, the smaller, so that the priority processing value can more accurately represent the current network operation status compared with the associated value, thereby identifying the alarm event more timely.

[0090] Preferably, the preset first weight and the second weight are 0.7 and 0.3 respectively.

[0091] In the present invention, the value of the preset first weight is greater than the value of the second weight, because the associated value needs to be used as the main reference, and the fluctuation change in the previous calculation cycle is used as an auxiliary reference.

[0092] Each type of operating parameter has a pre-set maximum value, which is generally the performance upper limit of each type of operating parameter. For example, for the operating parameter of bandwidth usage, it corresponds to the maximum available bandwidth of the CDN node.

[0093] Preferably, analyzing the operating parameters to determine whether a preset type of alarm event occurs includes:

[0094] The operating parameters are calculated with the alarm rules of various types of preset alarm events input to determine whether the number of alarm rules that the operating parameters meet is greater than or equal to 1. If so, it indicates that an alarm event of the preset type has occurred. If not, it indicates that an alarm event of the preset type has not occurred.

[0095] Since one type of operating parameter may correspond to different types of alarm events, for example, an operating parameter such as bandwidth usage may correspond to multiple types of alarm events such as excessive bandwidth usage and network congestion. Therefore, by comparing the operating parameters with the alarm rules of each type of alarm event, all existing alarm events can be determined.

[0096] The preset alarm rule can be to judge the value of the operating parameter. For example, for an alarm event of excessive bandwidth usage, a bandwidth usage greater than 90% of the maximum available bandwidth of the CDN node indicates that the alarm rule of the alarm event of excessive bandwidth usage is met. For an alarm event of network congestion, a bandwidth usage greater than 97% of the maximum available bandwidth of the CDN node indicates that the alarm rule of the alarm event of excessive bandwidth usage is met.

[0097] Preferably, updating the association value between the operating parameter and the alarm event includes:

[0098] An adaptive update cycle is used to update the correlation value between the operating parameters and the alarm events.

[0099] In the present invention, the update cycle is continuous. After the previous update cycle ends, the next update cycle begins and the calculation of the new association value begins. Before the new association value is calculated, the old association value is used to calculate the priority processing value.

[0100] Since the calculation amount of the association value is too large, the present invention does not use a fixed update period to calculate the association value, because this affects the efficiency of judging the operating parameters. Therefore, the present invention calculates the association value based on an adaptive update period, so that the update period of the association value can be adaptively changed with the change of the communication environment, thereby improving the effectiveness of the association value.

[0101] Preferably, the calculation process of the adaptive update period includes:

[0102] Using cycupd b,c+1 Indicates the c+1th update cycle corresponding to the running parameters of type b. After the cth update cycle ends, if numtrig c Greater than numtrig c-1 , then cycupd b,c+1 The calculation formula is:

[0103]

[0104] If numtrig c Less than or equal to numtrig c-1 , then cycupd b,c+1The calculation formula is:

[0105]

[0106] cycupd b,c Indicates the cth update cycle corresponding to the running parameters of type b, numtrig c and numtrig c-1 Indicates the number of alarm events triggered by the b-type operating parameters in the c-th and c-1-th update cycles, respectively. c It indicates the total number of b-type operating parameters obtained by the data acquisition module in the c-th update cycle, basic is the basic change duration of the update cycle, cycupdmi and cycupdma respectively indicate the minimum and maximum values ​​of the preset update cycle.

[0107] Specifically, the value of c is greater than or equal to 2. In the present invention, the update cycle can change adaptively with the change of the operating parameters in the first two update cycles, so that when the number of alarm times triggered by the operating parameters shows an increasing trend, the update cycle can be shortened so that an associated value that is more in line with the actual network operating environment can be obtained, and the alarm event can be monitored more timely. When the number of alarm times triggered by the operating parameters shows a decreasing trend, the update cycle is increased so that the degree of influence on the judgment efficiency of the operating parameters can be reduced. In addition, the degree of change of the update cycle of the present invention is also related to the number of alarm events triggered by the operating parameters of type b in an update cycle. The larger the number, the greater the degree of change of the update cycle. Therefore, the update cycle of the present invention achieves a balance between the degree of fit between the associated value and the actual network operating environment and the degree of influence on the judgment efficiency of the operating parameters, thereby providing customers with an alarm management system with better alarm performance.

[0108] When the alarm rules are met, an alarm event is triggered.

[0109] Preferably, the basic change duration of the update cycle is 2 hours.

[0110] Preferably, the minimum value and the maximum value of the preset update period are 12 hours and 36 hours respectively.

[0111] The minimum and maximum values ​​of the update cycle can be set according to the data update frequency of the CDN node. The higher the update frequency, the smaller the minimum and maximum values ​​are, so that the associated value can be more in line with the actual network operation situation.

[0112] Preferably, the first update cycle and the second update cycle lack data that can be calculated, so the first update cycle and the second update cycle are cycupdmi.

[0113] Preferably, updating the association value between the operating parameter and the alarm event includes:

[0114] For an alarm event of type d, the type of the benchmark operating parameter corresponding to the alarm event of type d is represented as type d ;

[0115] After the cth update cycle is over, use U b,c and Indicates the operating parameters and type of type b obtained in the cth update cycle d A collection of operating parameters of the type;

[0116] Will U b,c and The correlation coefficient between them is used as the association value between the operating parameters of type b and the alarm events of type d.

[0117] Specifically, each type of alarm event has a corresponding benchmark operating parameter, which is one of all types of operating parameters, and the benchmark operating parameter is the operating parameter with the greatest correlation with the alarm event. By analyzing the data features of different types of operating parameters when a type of alarm event occurs, the benchmark operating parameters of the event can be determined.

[0118] Preferably, the data features of different types of operating parameters are analyzed, and the analysis process includes:

[0119] Data preprocessing: First, the operating parameters are preprocessed, including data cleaning, formatting, and normalization, to ensure the accuracy of subsequent analysis.

[0120] Feature extraction: Extract data features from operating parameters, such as traffic change slope, request error rate, user access pattern, etc.

[0121] Statistical identification: Statistical identification of extracted features is performed using machine learning algorithms, such as cluster analysis, classification algorithms, and anomaly detection models. For example, through cluster analysis, the occurrence probability of various types of data features can be found when an alarm event such as excessive bandwidth usage occurs. The operating parameters of the type corresponding to the data feature with the highest occurrence probability are used as the benchmark operating parameters for alarm events such as excessive bandwidth usage.

[0122] In addition, the benchmark operating parameters of the alarm event can also be manually set according to industry experience. For example, for an alarm event of excessive bandwidth usage, the benchmark operating parameter is set to the bandwidth usage.

[0123] Preferably, if Figure 2As shown, the alarm module includes an alarm level acquisition unit and an alarm unit;

[0124] The alarm level acquisition unit is used to acquire the alarm level corresponding to the alarm event of the preset type that occurs;

[0125] The alarm unit is used to issue an alarm based on the alarm strategy corresponding to the alarm level.

[0126] Specifically, different types of alarm events correspond to different alarm levels. One setting method is that the greater the impact on the performance of the CDN, the higher the alarm level of the alarm event.

[0127] For example, an alarm event that has a greater impact on the cache hit rate has a higher alarm level.

[0128] The alarm level can be divided into 10 levels, and then the impact of different types of alarm events on the cache hit rate can be quantified. For example, the maximum percentage of the reduction in the cache hit rate can be used for quantification. An alarm level is set every 10%. When the reduction in the percentage of the reduction in the cache hit rate is in the interval [0, 10%], it is the alarm level 1. By analogy, the percentage interval of the reduction in the cache hit rate corresponding to each alarm level is obtained.

[0129] Another setting method is that the greater the economic loss caused, the higher the alarm level of the alarm event.

[0130] Preferably, the alarm is issued based on the alarm strategy corresponding to the alarm level, including:

[0131] In the present invention, a higher alarm level indicates a greater impact of the alarm event on the performance of the CDN, and each alarm level corresponds to a different alarm strategy.

[0132] Preferably, the alarm strategy includes an alarm method and an alarm content.

[0133] Alert methods include email alerts, phone alerts, pop-up alerts, push alerts, etc. When an alert is issued, the lower the alert level, the fewer different types of alert methods are used. The alert content is all types of alert events that occur within the set analysis period.

[0134] For example, when the alarm level is level 10, the alarm mode includes all types of alarm means.

[0135] Compared with the existing content distribution network alarm management method, the present invention updates the correlation value between the operating parameters and the alarm events at irregular intervals, and then calculates the priority processing value regularly. In this way, when the data analysis module receives multiple types of operating parameters within the set analysis cycle, it can give priority to analyzing and calculating the operating parameters with high priority processing values, thereby improving the timeliness of discovering alarm events.

[0136] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0137] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An alarm management system for a content distribution network, characterized in that: Including data acquisition module, data analysis module and alarm module; The data collection module is used to collect the operating parameters of the content distribution network; The data analysis module includes a calculation unit, an analysis unit and an update unit; The calculation unit is used to calculate the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event; The analysis unit is used to analyze the operating parameters based on the priority processing values, including: If the number of types of operating parameters received within the set analysis period is greater than or equal to 2, the operating parameters are analyzed in descending order of priority values ​​of the operating parameters to determine whether an alarm event of a preset type occurs; If only one type of operating parameter is received within the set analysis period, the operating parameter is analyzed to determine whether an alarm event of the preset type occurs; The updating unit is used to update the correlation value between the operating parameter and the alarm event; The alarm module is used to generate an alarm according to a preset alarm strategy when an alarm event of a preset type occurs.

2. The alarm management system for a content distribution network according to claim 1, characterized in that: Collect the operating parameters of the content delivery network, including: The operating parameters of each CDN node in the content distribution network are collected separately.

3. The alarm management system for a content distribution network according to claim 1, characterized in that: The priority processing value of the operating parameter is calculated based on the correlation value between the operating parameter and the alarm event, including: A preset calculation cycle is used to calculate the priority processing value of the operating parameter based on the correlation value between the operating parameter and the alarm event.

4. The alarm management system for a content distribution network according to claim 3, characterized in that: The calculation process of the priority value includes: Use b to indicate the type of running parameters, and use assval b Indicates the maximum value of the association value between the operating parameter of type b and various types of preset alarm events; The calculation formula for the priority processing value of type b operating parameters is: pripro b represents the priority value of the operating parameter of type b, α1 and α2 represent the preset first weight and second weight respectively, cyc b Indicates that in the time period [coltim b -H,coltim b ], the set of b-type operating parameters collected by the data acquisition module, coltim b Indicates the end time of the previous calculation cycle, H is the preset calculation cycle, and i represents cyc b The value of the element in, numi represents cyc b The number of elements in mai represents the maximum value of the operating parameter of type b.

5. The alarm management system for a content distribution network according to claim 1, characterized in that: Analyze the operating parameters to determine whether a preset type of alarm event occurs, including: The operating parameters are input into the preset alarm rules of various types of alarm events for calculation to determine whether the number of alarm rules that the operating parameters meet is greater than or equal to 1. If so, it indicates that an alarm event of the preset type has occurred. If not, it indicates that an alarm event of the preset type has not occurred.

6. The alarm management system for a content distribution network according to claim 1, characterized in that: Update the association values ​​between operating parameters and alarm events, including: An adaptive update cycle is used to update the correlation value between the operating parameters and the alarm events.

7. The alarm management system for a content distribution network according to claim 6, characterized in that: The calculation process of the adaptive update period includes: Using cycupd b,c+1 Indicates the c+1th update cycle corresponding to the running parameters of type b. After the cth update cycle ends, if numtrig c Greater than numtrig c-1 , then cycupd b,c+1 The calculation formula is: If numtrig c Less than or equal to numtrig c-1 , then cycupd b,c+1 The calculation formula is: cycupd b,c Indicates the cth update cycle corresponding to the running parameters of type b, numtrig c and numtrig c-1 Indicates the number of alarm events triggered by the b-type operating parameters in the c-th and c-1-th update cycles, respectively. c It indicates the total number of b-type operating parameters obtained by the data acquisition module in the c-th update cycle, basic is the basic change duration of the update cycle, cycupdmi and cycupdma respectively indicate the minimum and maximum values ​​of the preset update cycle.

8. The alarm management system for a content distribution network according to claim 7, characterized in that: Update the association values ​​between operating parameters and alarm events, including: For an alarm event of type d, the type of the benchmark operating parameter corresponding to the alarm event of type d is represented as type d ; After the cth update cycle is over, use U b,c and Indicates the operating parameters and type of type b obtained in the cth update cycle d A collection of operating parameters of the type; Will U b,c and The correlation coefficient between them is used as the association value between the operating parameters of type b and the alarm events of type d.

9. The alarm management system for a content distribution network according to claim 1, characterized in that: The alarm module includes an alarm level acquisition unit and an alarm unit; The alarm level acquisition unit is used to acquire the alarm level corresponding to the alarm event of the preset type that occurs; The alarm unit is used to issue an alarm based on the alarm strategy corresponding to the alarm level.

10. The alarm management system for a content distribution network according to claim 1, characterized in that: The alarm strategy includes the alarm method and alarm content.

Images